Clouds

Ssitzaedey
sities

ry
iy

mie 2
IER

:
7
os

peskins

7

Seon

~ a ~ in
(4 QUEEN’s UNIVERS)Ty OF BELFAST
SHELVED aT Oe 42]
CLAR
GEOSCIENCES LIBRARY -

6278790 [X]

UN ATIAAT YA

QUEEN'S UNIVERSITY BELFAST
Geography Department Librar
_ This book must be returned by

_ the latest date st

| : amped below,

| unless previously rehawee,

_ Otherwise a fine will he imnncad

_ QUEEN'S UNIVERSITY BELFAST

1 Geography Department Library”

| This book must be returned by

|the latest date stamped below,

| unless previously renewed, )
otherwise a fine will be imposed,

PLATE ¢.

elo uDs

A DESCRIPTIVE ILLUSTRATED GUIDE-BOOK TO THE
OBSERVATION AND CLASSIFICATION OF CLOUDS

BY

GEO. AUBOURNE CLARKE, F.R.P.S., F.R.Mzr.Soc.

WITH A PREFACE BY
siz NAPIER SHAW, LL.D., Sc.D., F.RS.

DIRECTOR OF THE METEOROLOGICAL OFFICE

LONDON

CONSTABLE ‘& COMPANY, LTD.
to-12 ORANGE STREET, W.C.
1920

We edna

M
ant
=

t

ENERODUOLORY NOTE.

Tus explanatory guide-book to cloud observation has
been prepared chiefly for those who are engaged in the
teaching of meteorology and of nature study, and also
for that numerous body of people to whose intelli-
gent interest in meteorology our country has always
been so deeply indebted. The book has therefore been
written as simply and as systematically as possible, the
subject-matter being grouped under different chapter-
headings in order that any particular branch of cloud-
study may be referred to immediately, and it is hoped
that any slight unavoidable overlap thereby caused may
be pardoned. Footnotes are added in all cases to render
reference to original papers easy.

For the professional meteorologist, the cloud-photo-
graphs themselves will doubtless hold the chief interest.
They represent a selection made from several hundred
pictures taken at Aberdeen by methods which were
varied so as to give the best possible representation of
each particular type.

I have to express my deep indebtedness to Sir
Mapieronaw, L,., sc.D.,; P.R.5., Director of, the
Meteorological Office, for writing a preface to the book,

Vv

vi INTRODUCTORY NOTE

and to the Meteorological Office for permission to incor-
porate the results of investigations made at Aberdeen
Observatory upon the clouds. To Captain C. K. M.
Douglas, late of the R.A.F., I am greatly obliged for
permission to reproduce a number of his wonderful
photographs of the clouds as seen from above, taken
from his aeroplane, and to him and also to Captain
A. E. M. Geddes, O.B.E., D.Sc., late of the Meteor-
ological Section, R.E., I have to express my thanks for
services rendered in reading the proofs, and for giving
valuable assistance.

GEO: AUBOURNES@ AACE.
THE UNIVERSITY OBSERVATORY,

Kinc’s COLLEGE,
ABERDEEN, December, 1919.

PREPACH

DY SIRENAPIER SHAW, LL.D.*Sc.D.,, F.R:S;

Iv is with much pleasure that I write some words of
introduction to this little book on Clouds, and for many
good reasons.

Since 1903 the author of it has been the “observer”
for the Meteorological Office at King’s College, Aber-
deen, where there is a meteorological observatory in
charge of a much-esteemed friend, Professor Niven. It
is one of seven established fifty years ago by the
Meteorological Committee of the Royal Society, then
newly appointed, for the purpose of obtaining continuous
records of the meteorological elements and auxiliary
eye-observations in order to provide material for the
scientific study of the sequence of weather represented
in outline by the daily synoptic charts compiled in the
Meteorological Office from telegraphic reports which
were initiated by FitzRoy in 1860. The new observa-
tories were modelled on the one developed at King
George III.’s Observatory at Richmond after it had
passed into the hands of the British Association for the
Advancement of Science in 1842. The other observa-
tories of the same type were at Glasgow, Stonyhurst,
Falmouth, Armagh, and Valencia. All of them, except
that at Valencia, which has retained its name though it

vil

Vili PREFACE

was removed to the mainland at Cahirciveen, were
placed in charge of some independent scientific authority,
and for the one at Aberdeen the Professor of Natural
Philosophy at King’s College was entrusted with the
administration of the sum of money set aside for its
maintenance.

The seven observatories formed the group of “ first
order stations” of the British meteorological network
when the collection of observations was systematised by
international agreement in 1874, and the chief of their
duties was to supply a series of hourly values of the
meteorological elements by which the data from the
climatological stations could be controlled.

As time went on these duties became a regular
routine: the records of pressure and temperature and
humidity were photographic, and what they had to tell
concerning the weather could only be known at inter-
vals of two days when the sheets were developed ; the
other records of wind, rain, and sunshine were daily.
Hence it came to pass that the ‘‘observer” had scanty
opportunity for tracing the connection between any un-
usual occurrence in the way of cloud or wind or rain
and the changes in the other elements with which it
was associated. Only two of the seven, Aberdeen and
Valencia, contributed observations:to the Daily Weather
Report, and so it came about that the certainty of get-
ting in due time good records of all the important
meteorological elements enabled the ‘ observers ” to lead
a regular and comparatively uneventful life tending
instruments and observing at fixed hours. The study
of weather as a personal enterprise degenerated into the

PREFACE ix

manipulation of the tabulations of records at a central
office.

But at Aberdeen Mr. Clarke has brought an official
nephoscope and his own camera to aid a very keen
eyesight for things to be seen in the sky and a most
valuable capacity for making sketches. This book is
a very welcome proof of the fact that the study of
weather still offers a field for individual enterprise even
when, and indeed because, the instrumental equipment
is of the best.

That is the first good reason, and I express it with
no little sense of gratitude to Mr. Clarke for his spon-
taneous activity and all that it has meant for the vitality
of the study of weather among those of us who have
had to work in the seclusion of Victoria Street and
South Kensington.

Secondly, in making selections from his store of
excellent photographs and sketches, Mr. Clarke has
been careful to fit them into the scheme of International
Classification. No one who makes such a collection can
be insensible to the failings of anybody else's classifica-
tion, and yet in such matters the better is very easily
the enemy of the good. We make more progress by
illustrating a classification that is generally accepted than
by devising a new one that in the first instance can only
form a subject of discussion.

Thirdly, there are the interest and the excellence of
the illustrations which Mr. Clarke has chosen and the
plain words with which he has introduced them. They
are full of suggestion, and, in spite of the transitory
nature of the fabric with which it deals, the book is a

is PREFACE

substantial addition to the growing edifice of the study
of weather.

Some years ago, when I had finished a book on
forecasting weather, I realised that it was incomplete
because, among other reasons, the study of clouds was
not included. Mr, Clarke’s book remedies that omis-
sion, and | will ask the gentle reader of that work to
regard it in that light. [do not wish to add anything
to what Mr. Clarke has said, and the only thing that I
notice that I should like to subtract is that at the top of
a cyclone air passes from low pressure to high pressure,
just opposite to what it does at the bottom. I know
many people think so, and some have said so, but
frankly I do not believe it does—the bottom is the
bottom and the top, if there is one, is the top, and not
a negative bottom.

NAPIER SHAW.

METEOROLOGICAL OFFICE,
toth January, 1920.

CONTENTS.

PAGE
Inrropuctory Notre ; : p ; : , . Vv
PREFACE . : ‘ 4 : ‘ . ; : : vil
SECTION I.
CHAP, ;
I. THE OBSERVATION OF THE CLOUDS. ; : ; 5
II. THE INTERNATIONAL CLASSIFICATION . : . : 8
III. CLoupD-FoRMS AND TRANSFORMATIONS . : ‘ eee
IV. Some CavusES OF THE FORMATION OF CLOUDS . Pee #
V. Cxoup Distripution, Heicuts, DIRECTIONS, VELOCITIES,
AND FREQUENCIES : ; ; f F : as
VI. THe AssocIATION OF CLOUD WITH WEATHER-TYPE . 83
SECTION II.
PLATES OF CLOUD PHOTOGRAPHS WITH DESCRIPTIVE
EXPLANATIONS. : : . ; : yes

INDEX . ; z . 4 . : ‘ d sae

Oo
ies)

xi

LIST OF ILLUSTRATIONS:

PLATES IN COLOUR.

- The structure of a nimbus cloud as revealed by search-

light eo Pe ; : , : : frontispiece
FACING PAGE

2. Optical phenomena in cirro-stratus , : ; “422
3. Lenticular cloud-banks at sunset. : : : ote
4. Band of cumulus in a line-squall . : : es

Io,

5

12.

PLATES IN BLACK AND WHITE.

A. Cirrus tufts A
B. Cirrus tufts, further development{ — ‘ ; ae
A. Cirrus, tufted parallel threads

B. Cirrus, tufted curved threads 97
A. Cirrus, in dense masses 3
B. Cirrus, changing into Ci.-cu.f — : ‘ ear
A. Cirrus, threads with nuclei } eo
B. Cirrus, detail in a long band

A. Cirrus, at two levels 1 ee
B. Cirrus, ditto, further development J ;

A. Cirro-stratus, developing from Ci. a
B. Cuirro-stratus, a dense layer ;

A. Cirro-nebula and solar halo —
B. Cuiro-stratus, with Ci. and Ci.-cu. f - . 7
A. Cirrus to Ci.-cu., intermediate type ) es
B. Cirrus, rippled type ' 7

xiii

xiv

18.

19.

20.

21.

22.

22

24.

Bs.

26.

24)

28.

> wb

WP WP p> WP WP

UP oP DP DP DP OP DP Ww > WP

LIST OF ILLUSTRATIONS

. Cirro-cumulus, “speckle-cloud ” \

Cirro-cumulus, with edging of Ci.

. Cirro-cumudus, in uniform sheets
. Cirro-cumulus, further development

. Cirro-cumulus, globular type }

Cirro-cumulus, waved type

Cirro-cumulus, in ‘‘ rosette” formation
Cirro-cumulus, sharply defined bands

. Alto-cumulus, in straight waves
. Alto-cumulus, in curved waves

. Alto-cumulus, in globular masses
. Alto-cumulus, in flakes j

. Alto-cumulus, typical form
. Alto-cumulus, in closed sheet J

: Cirro-cumulus, fused lenticular banks \

Cirro-cumulus, rippled lenticular bank

. Cirro-cumulus, heavy lenticular banks )

Cirro-cumulus, a large lenticular sheet {

Alto-cumulus, \enticular cloud-banks
Alto-cumulus, \enticular cloud-banks

False cirrus, above Cu.-nb.
False cirrus, becoming A.-cu.

Alto-cumulus castellatus, “ turret-cloud ”
Alto-stratus, typical form

Strato-cumulus, typical form }

. Strato-cumulus, heavy typical form

. Strato-cumulus, high type
. Strato-cumulus, in sinuous waves

. Strato-cumulus, wave-system forming |
. Strato-cumulus, long parallel bands {

Cumulus, small detached. clouds |
Cumulus, large typical cloud = f

FACING PAGE

104

105

106

107

imeys)

109g

IIo

ELT

riz

113

II4

11s

116

Loy,

118

119

29.

30.

can

22.

esr

34.

35:

ZO:

37:

38.

39-

40.

n - wW

LIST OF ILLUSTRATIONS

. Cumulus, in parallel lines
. Cumulus, a long cloud-bank

Cumulus, with fracto-cumulus )
Cumulus, becoming St.-cu. f ~
Cumulo-nimbus, thunder-cloud with “anvil”
Cumulo-nimbus, thunder-cloud without “anvil” {
Cumulo-nimbus, hail-squall cloud
. Cumulo-nimbus, shower-cloud
Cumutlo-nimbus, thunder cloud forming
Mammato-cumulus
. Stratus, typical form
. Stratus, in thin waves
LVimbus, a shower falling
Nimbus, with a rainbow
. Line-squall cloud, phase 1

Line-squall cloud, phase 2
. Line-squall cloud, phase a
. Line-squall cloud, phase 4
. Cumulo-nimbus, seen from eet

Strato-cumulus layer, from above

Strato-cumulus, billows on upper surface |

WP OP UP ob ob > Db Db ob Wb > WP

. Strato-cumulus, showing “‘cliff-front” —j
Rippled cloud-band \
Cumulus bank, seen from above

FIGURES AND DIAGRAMS IN TEXT.

. Diagrammatic representation of cloud-heights .
. Radiant-point of curved bands
. Radiant-point of straight bands
. Diagram of optical phenomena

. Formation of lenticular cloud-banks

XV

FACING PAGE

I20

122

123

Xvi

LIST OF ILLUSTRATIONS

. Development of Cumulo-nimbus “ anvil ”

. Air-motion in high and low-pressure systems
. Air trajectories in a depression

. Adiabatic and average lapse-rates

. Lapse-rates and condensation

. Air-motions in a line-squall

. Meteorological records during a line-squall

. Sectional diagram of a thunder-cloud

Frequencies of cloud-directions at Aberdeen

. Cloud distribution at meteorological stations
. Examples of weather-types

. Examples of weather-types

PAGE
40
56
aH

59
60

62
64
67
78
81
85
88

7

‘ os ‘ a

~ ¥
=
P - Cae
a a
© — - A
~
a
°
-

dil

3554 aa! Te et

CHARTER: |
THE OBSERVATION OF THE CLOUDS.

Croup observation must have commenced at a very
early period in the history of mankind. It is only when
we consider how deep an impression must have been
made upon the mind of early man by the gorgeous hues
of sunset or by the gloom and terror of the thunderstorm,
that we fully realise how natural it was for him to
associate such glories and such terrors with a deity who
had to be propitiated. Throughout the whole range of
pagan mythology we meet with gods of the sky and air,
of light and darkness, and of storm and thunder; gods
who were believed to control] the rain and deluge, whose
beneficence was shown by the sending of fertilising
showers, and whose kindling anger materialised in hail
and thunderstorms. One of the most effective of the
apotheoses in the old Norse mythology is that of Frigga,
the wife of Odin and goddess of the clouds, who is
pictured dwelling in Fensalir, the Hall of Mists, where
she employed her time with wheel and distaff spinning
the golden threads which were woven into bands of
brightly coloured cloud.

The references that are made to the clouds in the
Book of Job.and the Psalms prove that, in Biblical times,
these wonderful appearances in the heavens had aroused
among the Jews a spirit of admiring and reverential
contemplation, a spirit ae pes lasted through the ages

3

4 CLOUDS

in all lands, and which has been the inspiration of much
that is beautiful both in poetry and in prose, as well as
in painting. The early half of the nineteenth century saw
the genius of Turner applied in art to the delineation of
magnificent skies—skies in which the cloud /ovms as
well as their glories were for the first time correctly
rendered—and it also saw with what reverence and in
what glowing language the clouds could be described
by John Ruskin in the chapters devoted to them in >
‘Modern Painters”.

The same period was likewise noteworthy for the
fact that for the first time in this country an attempt was
made on scientific lines to classify clouds according to
their forms. There can be no doubt whatever that the
recognition of certain cloud-formations must have been
almost coeval with man’s first observations of the skies.
In particular the shepherds of the plains and the tillers
of the soil must have come to recognise that one form
of cloud brought to them rain while other forms did not ;
and they that “went down to the sea in ships” must
soon have become familiar with clouds that gave them
windy weather, and with others that presaged coming
storms. Shakespeare, in ‘Richard II,” says

Men judge by the complexion of the sky
The state and inclination of the day,

and the whole of our household words and maxims
about the weather, such as ‘“‘a red sky at night is the
shepherd's delight,” are simply the natural outcome of
man’s powers of correlating certain facts with certain
appearances of the sky.

The fact that cloud-forms fell into three broadly
distinct classes was the basis upon which Luke Howard

THE OBSERVATION OF THE CLOUDS 5

in 1803 suggested the first definite classification of clouds.
He recognised, as almost any observer of the heavens
could scarcely fail to do, that sometimes the clouds were
very light and wispy, spread in lines and patches with
a texture resembling hair ; at others they were heaped
up into mountainous masses; while a third form was
that of a more or less dense sheet or layer of cloud resting
on the ground—what we, in fact, call a fog. To the
first type he gave the name of c¢vrus (from the Latin word
meaning “a wisp of hair”); to the second he applied
the name of cumulus (meaning “a heap”); while the
third was called stvatus, because it was spread out in
a flat sheet. It is upon these three type-names that all
subsequent classifications have been built up. Naturally
there are many forms of cloud intermediate between
the main types above-mentioned—Howard himself re-
cognised and gave names to several—and it is precisely
this fact which has given rise, since Howard’s time, to
quite a number of different classifications. Prominent
among the more recent of these schemes of classifying
cloud-forms is that suggested by the Rev. W. Clement
Ley in his book “ Cloudland” which appeared in 1894.
In the same year a meeting of the International Meteoro-
logical Committee, held at Upsala in Sweden, entrusted
to Prof. H. H. Hildebrandsson of Upsala, M. L.
Teisserenc de Bort of Paris, and M. A. Riggenbach of
Zurich, the publication of an Atlas wherein, in accord-
ance with the resolutions adopted at a previous meeting
in Munich in 1891, the cloud-forms should be clearly
defined and illustrated. The Atlas was published shortly
afterwards, and since then a second edition has appeared
in 1910, This classification, which is based upon the
suggestions of the Hon. Ralph Abercromby and Prof.

6 . CLOUDS

Hildebrandsson, is still in use at Meteorological Stations
in-all countries, and thus serves the very useful purpose
of maintaining uniformity in cloud observation through-
out the world. The bases of its nomenclature are the
heights and forms of the clouds, and ten types are recog-
nised and are given definite names ; in addition to which
there are four qualifying adjectives applicable to certain
forms. ° The descriptions given and the pictures shown
are of very typical character, such indeed as are not likely
to be met with at all frequently by observers ; and this
fact has been instrumental in calling forth further sug-
gestions, mainly tending towards a multiplication of
types. In his beautifully illustrated book ‘Cloud
Studies,” which appeared in 1905, A. W. Clayden has
divided each of the International cloud-types into several
sub-types. But this procedure, in addition to making
the classification unwieldy, is open to the further objec-
tion that, particularly in the case of cirrus, the same
cloud-band may contain several of these sub-types within
its length simultaneously, and, owing to the rapid inter-
nal changes which -often characterise these clouds, one
sub-type may be transformed into another and then
perhaps return to its original form all within the space
of afew minutes. In fact, it may be said that so endless
are the possible varieties of the clouds, and so bewilder-
ing their changes of form, that no classification which
was based upon form alone could ever hope to be
complete, and it would be advisable rather to aim at a
further simplification of the present classification than to
extend it.

It may be that in the near future, the work that has
been done in the way of investigating the upper air and
clouds by aeroplane will provide us with a new basis

THE OBSERVATION OF THE CLOUDS 7

whereon we may re-classify our clouds, possibly accord-
ing to their physical relationships, but meantime, as the
International Classification is_ still regarded as the
standard one for observers, it will be well to set it forth
at length in the next chapter.

CHAGLE RAL
THE INTERNATIONAL CLASSIFICATION.

Tue International Classification divides clouds into
two main sections according to whether (a) they are
detached clouds with rounded upper outlines, types that
are most frequent in dry weather, or whether (4) they
are clouds of great horizontal extent, suggesting a layer
or sheet, the form most common in wet weather. Then
again it divides the clouds into upper, intermediate, and
lower classes according to their heights. Classes are set
apart for clouds which are known to be caused by diurnal
ascending currents, because these clouds have special
characteristics of their own, such as great vertical depth ;
and also for high fogs, which are really the lowest of
clouds.
The complete scheme then stands thus :—
A. Upper Clouds, average altitude go000 metres
(about 30,000 feet).
(2) te Cirrus:
(6) 2. Cirro-stratus.
B. /ntermediate Clouds, between 3000 and 7000
metres (10,000 and 23,000 feet).
(a) 3. Cirro-cumulus.
4. Alto-cumulus.
(4) 5. Alto-stratus.
C. Lower Clouds, below 2000 metres (7000 feet).
(2) 6. Strato-cumulus.
(6) 7. Nimbus.
(8)

THE INTERNATIONAL CLASSIFICATION 9

D. Clouds of Diurnal Ascending Currents.

(z) 8. Cumulus: height of base 1400 metres
(4500 feet), top 1800 metres (6000 feet).

(4) 9. Cumulo-nimbus : height of base 1400 metres
(4500 feet), top 3000 metres to 8000 metres
(10,000 feet to 26,000 feet).

E. High Fogs, under 1000 metres (3000 feet).

10. Stratus.

The definitions and descriptions given of the cloud-
forms are as follows, the letters within brackets after the
cloud names being the recognised contraction for the
name :—

1. Cirrus (Cz). Detached clouds of delicate and
fibrous appearance, often showing a feather-like structure,
generally of a whitish colour.—Cirrus clouds take the
most varied shapes, such as isolated tufts, thin filaments
on a blue sky, threads spreading out into the form of
feathers, curved filaments ending in tufts ; they are some-
times arranged in parallel belts which cross a portion of
the sky in a great circle, and by an effect of perspective
appear to converge towards a point on the horizon, or,
if sufficiently extended, towards the opposite point also.
Cirro-stratus and Cirro-cumulus are also sometimes

arranged in similar bands.
2. Cirro-stratus (Ct. St.). A thin whitish sheet of
clouds sometimes covering the sky completely and giving
it only a milky appearance (it 1s then called Cirro-nebula),
at other times presenting, more or less distinctly, a for-
mation like a tangled web.—This sheet often produces
halos around the sun or moon.

3. Cirro-cumulus (Ct. Cu.), Mackerel Sky. Small
globular masses or white flakes without shadows or

10 CLOWGSS

showing very slight shadows, arranged mm groups and
often im lines.

4. Alto-cumulus (A. Cu.). Larger globular masses,
white or greyish, partially shaded, arranged in groups or
lines and often so closely packed that their edges appear
confused—The detached masses are generally larger
and more compact (resembling Strato-cumulus) at the .
centre of the group but the thickness of the layer varies.
At times the masses spread themselves out and assume
the appearance of small waves or thin slightly curved
plates. At the margin they form into finer flakes (re-
sembling Cirro-cumulus). They often spread themselves
out in lines in one or two directions,

5. Alto-stratus (A. St.). A thick sheet of a grey or
bluish colour, sometimes forming a compact mass of dark
grey colour and fibrous structure.—At other times the

‘sheet is thin, resembling thick Cirro-stratus, and through

it the sun or moon may be seen dimly gleaming as
through ground glass. This form exhibits all changes
peculiar to Cirro-stratus, but from measurements its
average altitude is. found to be about one-half that of
Cirro-stratus.

6. Strato-cumulus (St. Cu.). Large globular masses
or rolls of dark clouds often covering the whole sky, es-
pecially in winter.—Generally Strato-cumulus presents
the appearance of a grey layer irregularly broken up into
masses of which the edge is often formed of smaller
masses, often of wavy appearance resembling Alto-
cumulus. Sometimes this cloud-form presents the
characteristic appearance of great rolls arranged in
parallel lines and pressed close up against one another.
In their centres these rolls are of a dark colour. Blue
sky may be seen through the intervening spaces, which
are of a much lighter colour. (Roll-cumulus in Eng-

THE INTERNATIONAL CLASSIFICATION 11

land, Wulst-cumulus in Germany.) Strato-cumulus
clouds may be distinguished from Nimbus by their
_ globular or rolled appearance, and by the fact that they
are not generally associated with rain.

7 Nimbus (N6.), Rain clouds. A thick layer of
dark clouds without shape and with ragged edges, from
which steady rain or snow usually falls—Through the
openings in these clouds an upper layer of Cirro-stratus or
Alto-stratus may be seen almost invariably. If a layer of
Nimbus separates up in a strong wind into shreds, or if
small loose clouds are visible floating underneath a large
Nimbus, the cloud may be described as /vacto-Nimbus
(Fr. Nb.), (“Scud ” of sailors).

8. Cumulus (Cu.), Wool-pack clouds. Thick clouds
of which the upper surface rs dome-shaped and exhibits
protuberances while the base ts horizontal, — These
clouds appear to be formed by a diurnal ascensional
movement which is almost always noticeable. When the
cloud is opposite the sun, the surfaces facing the observer
have a greater brilliance than the margins of the protuber-
ances. When the light falls aslant, as is usually the case,
these clouds throw deep shadows ; when, on the contrary,
the clouds are on the same side of the observer as the
sun, they appear dark with bright edges. True Cumu-
lus has well defined upper and lower margins, but in
strong winds a broken cloud resembling Cumulus is
often seen, in which the detached portions undergo con-
tinual change. This form may be distinguished by the
name /’yvacto-cumulus (Fr. Cu.).

9. Cumulo-Nimbus (Cu. Nb.), the Thunder-cloud ,;
Shower cloud. Heavy masses of cloud rising in the form
of mountains, turrets, or anvils, generally surmounted by
a sheet or screen of fibrous appearance (false Cirrus),
and having at its base a mass of cloud similar to Nimbus.

ro CLOUDS

—From the base local showers of rain or snow (oc-
casionally of hail or soft hail) usually fall. Sometimes
the upper edges assume the compact form of cumulus,
and form massive peaks round which delicate ‘false
cirrus” floats. At other times the edges themselves
separate into a fringe of filaments similar to cirrus
clouds. This last form is particularly common in spring
showers. The front of thunder-clouds of wide extent
frequently presents the form of a large arc spread over
a portion of a uniformly lighter sky.

10. Stratus (St.). A uniform layer of cloud re-
sembling a fog but not resting on the ground.—When
this sheet is broken up into irregular shreds in a wind,
or by the summits of mountains, it may be distinguished
by the name /’racto-stratus (Fr. St.).

To the above definitions are appended a few qualify-
ing remarks upon certain variations likely to be met
with when obsérving the clouds. These are -—

(a) During summer all low clouds tend to assume
forms resembling cumulus. In such cases they should
be noted as Stratus or Nimbus cumulifornis.

(4) At times a cloud may present a mammillated
lower surface. This appearance should be noted under
the name of Mammato-cumulus.

(c) The ovoid form with sharp edges assumed. by
certain clouds, particularly during the occurrence of
strocco, mistral or fohn, should be designated /enticularis.
For example, Cumulus lenticularis, Stratus lenticularis.
Such clouds frequently show iridescence.

The distribution of the clouds according to the
heights assigned to them in the International Classifica-
tion is shown in diagrammatic form in Fig. 1. The
diagram is intended to represent the heights of three
well-known mountains; in the background is Mount

NY }
‘veg
a i ty wt 44
\ 4 ‘i va ihe
ye: 4,

%
—)
=
a |
=
=
oS

'
S
f=
~J
“8

i
h)

inal =

SS

SSss

2Ss—:
Ss

\ M)

U
NX

a
]

= § ROR i
SS —— : 2 =

CIRRO- STRATUS

134 PUBSNOY]

—
Lae)

Fic, 1.—Diagrammatic representation of cloud-heights.

14 CLOUDS

Everest, 29,000 feet high, in the middle distance stands
Mont Blanc, 15,800 feet high, while in the foreground
is the highest mountain in the British Isles, Ben Nevis,
4400 feet in altitude. Across these mountain peaks,
at the levels where they are on the average found, are
shown the various cloud-types, the cirrus and cirro-
stratus just level with the summit of Mount Everest,
and Ben Nevis below all but the stratus. There is, of
course, no clearly-defined level to each type such as is
shown iri the diagram, because in actuality one cloud-
type merges gradually into another ; cirro-stratus may
be found as much lower than its average height as
alto-stratus may be found higher than its own normal
level; while nimbus may at times be found no higher
than stratus. The diagram is so arranged that clouds
which show forms intermediate between them are placed
vertically over each other ; thus cirrus often gives place
to cirro-cumulus, which in turn may become alto-cumulus,
while the latter frequently develops into strato-cumulus.

Perhaps the most striking feature of the diagram is
the relative size of the cumulo-nimbus; the drawing
shows, far better than any description, how very unique
this type of cloud is when well developed. Other cloud-
layers may at times be one or two thousand feet thick,
but a cumulo-nimbus with its base no higher than four
thousand feet may tower upwards to fifteen or even to
twenty thousand feet, thus becoming an actual mountain
of condensation.

The vertical shading below nimbus and cumulo-
nimbus is intended to represent falling rain, and the
anvil-shaped addition at the top of cumulo-nimbus repre-
sents the “false cirrus” which frequently accompanies
these clouds.

CHAI Di 111
CLOUD-FORMS AND TRANSFORMATIONS.

Ong of the chief attributes of Cz7rus cloud is its liability
to appear arranged in long bands. Very often after a
spell of two or three days of cloudless blue sky the first
clouds to appear on the southern horizon are a few faint
threads of delicate white cirrus. As time goes on these
threads increase in quantity until they spread themselves
right across the sky in a series of bands which seem to
proceed from one particular point on one horizon, and
to converge again towards another point diametrically
opposite the first one. The bands are, of course, strictly
parallel to each other, but the fan-like arrangement is
due to the effect of perspective. Those who are not
familiar with the laws of perspective will find them very
clearly explained and demonstrated by Ruskin in his
chapters on clouds in ‘‘ Modern Painters”*; and their
effect upon the appearance of cloud-bands is shown in
Figs. 2 and 3 in the present work. Not only cirrus
but practically all the detached clouds from strato-
cumulus upwards exhibit this banded arrangement oc-
casionally ; and in the case of cloud-layers which are
striated or waved along the lines of divergence, the
effect of perspective is to create widely-differing appear-
ances in the bands according to the apparent inclination
of the bands to the vertical. For example, in Fig. 2

1], Ruskin, “ Modern Painters,” Vol. V, Part VII, chap. il.
(15)

ST oe ee Ss

Fic. 3.—Radiant-point : effect of perspective on straight bands with
cross-Striations,

CLOUD-FORMS AND TRANSFORMATIONS 17

there is shown a very idealised example of strato-
cumulus, where gently but regularly waved bands are
separated by spaces of sky of exactly the same width as
the bands themselves. The bands are diverging from
a point on the horizon exactly on the right-hand margin
of the picture, and it will be seen at once that the un-
dulations of the bands become more cramped together
the nearer they are to the horizon, and also the more
nearly the bands approach the vertical. In Fig. 3 there
are shown some bands of cirrus having a feather-like
structure, radiating from a point nearer to the centre
of the picture. In this case the effect of perspective is
to alter very considerably the angle which the curved
filaments make with the band itself. Several still more
complicated effects are given by Ruskin in the work
referred to. A knowledge of these laws is very useful,
because an observer familiar with them is able to trace
this “ Radiant-point,” as it is called in meteorology—or
‘““Vanishing-point,” as it is called in perspective—even
when the cloud-bands are considerably broken up and
rather irregular.

When the bands of cirrus are examined in detail it
is amazing to see what a variety of intricate yet orderly
structure they show. Sometimes, indeed, they consist of
a multitude of the very finest of thread-like fibres which
singly would be almost invisible, but which when grouped
together give a somewhat indefinite and hazy appearance
to the whole band ; this is the form which so readily
coalesces into a sheet and develops into cirro-stratus
(Plate toa). But much more frequently the cirrus
bands are composed of small filaments seemingly inex-
tricably tangled, or of little tufts with tapering extremities,

sometimes called “ tufted” or “ tailed” cirrus. Occasion-
2

18 CGUDS

ally a regular series of these tufts may have their “ tails ”
lying in parallel curves which unite into one long band,
the whole arrangement then resembling a gigantic feather.
Another recurring form is that wherein a strongly defined
thread becomes a core round which are grouped many
fainter filaments ; this form frequently assumes a rippled
appearance strongly resembling small wavelets on water
looked at from above (Plate 128). This last form is
called “ change cirrus ” by Clayden,' and the term is very
apt, for the cloud seems to be in a state of continual flux,
changing from thread to ripple and back again in the
course of a few moments. In fact, all the above-men-
tioned forms, tufts, plumes, and wavelets seem curiously ,
liable to interchange their patterns, and are often found
simultaneously in the sky. Usually they eventually de-
velop into the higher forms of cirro-cumulus. There is,
indeed, a type which seems to partake more of the
character of cirro-cumulus than of cirrus, where the deli-
cate threads seem to have become gathered up into
intensely white little balls or patches (Plate 12 a).

At certain times, not only may cirrus be seen to de-
velop into cirro-cumulus, but there may also be found both
types together in the same cloud. Some examples of this
are shown in Plates 13 4 and B; and in connection with
this fact an interesting point has been raised by Captain
Cave,’ who has observed these clouds in the south of
England. True cirrus cloud is generally regarded as
being composed of tiny ice-crystals while cirro-cumulus
is believed to be composed of water-droplets which may

1A, W. Clayden, ‘ Cloud Studies,” p. 37.
* Capt. C. J. P. Cave, R.E., ‘‘The Forms of Clouds,” Quarterly

Journal of the Royal Meteorological Society, January, 1917. Vol.
XLII, p. 66.

\

CLOUD-FORMS AND TRANSFORMATIONS 19

be supercooled, but nevertheless still liquid’ Captain
Cave remarks that it is an easy matter to understand
_ that at such heights, where the temperature is far below
freezing-point, cirro-cumulus cloud might very readily
become transformed into cirrus by the freezing of the
water-droplets ; but it is not so easy to understand how
the ice-crystals of cirrus can change into the water-
droplets of cirro-cumulus. Yet the latter transformation
has* been observed at Aberdeen far more frequently than
the former, and the fact still requires explanation.

Cirro-stratus.—Cirrus clouds, it has been said, are
composed of tiny ice-crystals, and when the cirrus has
become cirro-stratus, some very beautiful appearances
are produced in the heavens through their agency by the
light from the sun and moon. The tiny ice-crystals
exist in the form of hexagonal thin plates and needles,
and act like prisms, refracting the rays of light that fall
upon their faces. Thus are produced a series of optical
phenomena, the more common of which are indicated
in the diagram, Fig. 4.

The phenomenon most frequently observed is the
“halo of 22°” as it is termed (a in illustration). This is
a circle of brilliant light seen round the sun or moon at
an angular radius of about 22°. The circle, in the case
of the sun, has its inner edge tinged with red quite
distinctly, and the outer edge may occasionally show a
faint tinge of greenish-yellow, but usually this colour is
liable to be merged into the general milky tint caused
by the white cirro-stratus overspreading the blue of the
sky. When the moon is the source of light, the halo is
most frequently colourless, but when an exceptionally
fine lunar halo is observed the inner edge of the ring is
of a dull orange-red tint. Occasionally, but much more

CLOUDS

20

‘euauousyd jeondo awos jo weiseiq—? ‘O17

CLOUD-FORMS AND TRANSFORMATIONS 21

rarely, there is seen a larger halo, of 46° radius (6 in
figure), but this ring is much less bright than the
smaller halo, and usually colourless. If colour is present
the inside edge of the ring is red, as in the case of the
smaller halo. Both of these halos may have at the top
of the circle and in contact therewith portions of other
circles which are called arcs of contact (c¢ ¢ in figure).
These arcs are usually very bright where they unite
with the halo and they likewise have the red colour on
the edge nearest the sun. It is not by any means un-
common to find the arcs of contact visible even though
there be no halo apparent, in fact the arc of contact to
the large halo of 46° is more frequently seen than is the
halo itself, probably because this arc of contact is often
very brilliant and strongly coloured, much more so,
indeed, than even the smaller halo is. In the coloured
picture (Plate 2) this arc is shown without the halo of
46 being present. Arcs of contact are not limited to
‘the upper part of the halo circles ; they may be formed
also at the lower parts, and even-at the sides of the halos,
but are very seldom seen. In addition to these two
halos there is sometimes seen a white ring passing
through the sun and running right round the sky parallel
to the horizon (d in figure). It is generally termed the
mock sun ring, because on it at several points may be
seen much brighter spots of light, which have been named
mock suns. Usually there is found one, called the
counter sun, situated diametrically opposite the actual
sun, and sometimes two others at about 60° along the
ring on either side of this counter sun.

Mock suns and mock moons (or parhkefa and parase-
Jen@ as they are named) may often be seen even though
the horizontal ring is absent. They are most generally

6

22 CLOUDS

found just outside the halo of 22°, at the points where
the horizontal ring if present would meet the halo (e in
figure), and very frequently they may be seen when no
halo at all is present. They are as a rule blindingly
bright and strongly coloured, red being, as in the case
of the halos, nearest the sun. The most brilliant parhelia
seem to be formed in a coarser or lower type of cirro-
stratus than that which gives rise to the other halo phe-
nomena, and the same may be said of another optical
phenomenon called the sux pillar (or moon pillar, as the
case may be). This is a vertical pillar or band of light,
best seen when the sun is very low in the heavens; it is
usually of a bright white or golden colour, though oc-
casionally an orange-red one has been seen when the sun
has just disappeared below the horizon.

The ordinary halos and ares of contact are, on the
other hand, best displayed when the cirro-stratus is very
thin and without texture. There are occasions when the
whole sky appears of a pale milky or silvery-blue tint,
and it would hardly be expected that any cloud were
present, but a very brilliant halo demonstrates the
presence of a layer of ice-crystals. This form of cirro-
stratus is called “cirrus haze,” or “cirro-nebula,” and is
frequently the first stage in the formation of the ordinary
type of cirro-stratus, which latter is in this event very
uniform in appearance. Sometimes, however, the cirro-
stratus shows distinct detail, such as a matted appear-
ance due to the interlacing of twisted fibres, or else a
beautiful web-like structure formed by parallel threads
lying in two directions; in this latter case it is not
uncommon to find in places a denser warp running
through the web. Another form is a rather coarse floc-
cular arrangement which really seems to be a stage of

PLATE 2.

CLOUD-FORMS AND TRANSFORMATIONS 23

development intermediate between cirro-stratus and
cirro-cumulus.

Owing to the circumstances attending its formation
cirro-stratus is seldom a cloud of long duration, it is
usually the matter of only a few hours before it has de-
veloped into, or given place to, the denser and lower
alto-stratus ; this is because both these cloud-types are
associated with depressions, and therefore with rapidly
increasing condensation. On the other hand, when
cirro-stratus does occur it is very widespread, frequently
covering more than half the total area of a depres-
sion.

Cirro-cumulus.—Cirro-cumulus is pre-eminently the
cloud of beauty. The wonderful regularity shown in
the arrangement of its bands and masses, the delicacy
of the little cloudlets themselves, and the vastness of
their numbers, all combine to invest this cloud with a
distinctiveness that compels our admiration.

The close association of cirrus with cirro-cumulus has
already been referred to. When an actual transforma-
tion of the former into the latter does take place, the
cirro-cumulus thus formed is almost always of a very
delicate type, the cloudlets being so small as hardly to
be visible individually, and generally of an almost trans-
parent whiteness. It has appropriately been named Cz7~
ro-macula or speckle-cloud by Clement Ley (Plate 13 4).

Sometimes all sizes of cloudlets may be seen together
in one sheet—those at the edges being the smallest—so
that the size of the cloudlet is not necessarily an actual
indication of the relative height of the cloud. One
rather curious variety is occasionally seen wherein the
cloudlets seem to be compound, a series of rather diffuse
cloudlets being grouped into an irregular ring, thus

24 CLOUDS

giving a “rosette” form and rather large size to the
cloudlets (Plate 16 4).

The waved forms of cirro-cumulus are equally as
beautiful as the globular ones, and when well-developed
are even more striking in appearance (Plate 168). The
waved form appears to be more commonly met with
when the cirro-cumulus is arranged in long bands, while
the globular type is usually found in irregular patches or
sheets of cloud, but even in the latter case there is often
to be distinguished an undulatory arrangement in one
direction or even in two directions which are frequently
nearly normal to each other (Plate 15 a).

Cirro-cumulus appears, from observations at Aber-
deen, to be a cloud of remarkably rapid formation, and
the same remark has been made by Clayden,' who has
observed them in Devonshire. Ina clear blue sky there
may suddenly appear and rapidly develop patches and
extensive sheets of a uniform dazzlingly white cloud
resembling cirro-stratus, but much whiter and more
solid-looking, and after the lapse of some little time they
will be seen to break up at the edges into flakes or
globular cloudlets of the cirro-cumulus type, and before
very long the whole sky may be overspread with a layer
of ordinary cirro-cumulus. One such development seen
at Aberdeen commenced about 9 a.m. and was not com-
plete till nearly three hours later, but this was an excep-
tional case ; it is usually a matter rather of minutes than
of hours for the development to become complete.

In cloud-sheets composed of cirro-cumulus, and also
in those formed of the higher and thinner varieties of
alto-cumulus, there may be seen round the sun or moon

Clayden, ‘‘Cloud Studies,” p. 57.

CLOUD-FORMS AND TRANSFORMATIONS 2s

a strongly coloured circle, which is almost always of
considerably smaller diameter than the halo seen in
cirro-stratus. It is caused by the diffraction of the rays
of light by the small particles of the cloud, which latter
has therefore been argued to consist of minute water-
drops, and not of ice-crystals as in the case of the cirro-
stratus, The colours of the corona, as this ring is termed,
are very evident, and follow in the reverse order from
those of the halo. In the corona violet is nearest the
sun and is followed by the other colours of the spectrum,
blue, green, yellow, orange, and red, the last-named being
on the outside of the ring. Sometimes, when the con-
ditions are favourable, a second, or even more frequently
repeated ring, showing the colours generally from green
to red, may be seen following outwards and beyond the
first red, thus producing a double, or multiple, corona.
Immediately within the violet of the first corona there
is a brownish-coloured ring which fades gradually into
the general whiteness that fills up the space between
the coronal ring and the sun (or moon). On a clear
night in winter-time, when the moon is high in the
heavens, a faint corona is occasionally seen without any
cloud being apparent, but this fact is indicative of the
presence of water-droplets in the atmosphere, and cirro-
cumulus is almost certain to appear before long.
Alto-cumulus.—The distinguishing feature between
cirro-cumulus and alto-cumulus, according to the Inter-
national Classification, is that of apparent size and of the
liability of the latter cloud to show shadows. But the
heavier types of the former cloud do occasionally have
faint shadows upon them, especially when densely packed,
while the latter cloud, though larger in appearance, is
often so thin as to be totally devoid of shadow-detail,

0 CLOUDS

All the various forms of cirro-cumulus, globular, flaked
and waved, are likewise to be found in alto-cumulus, and
the last-named skies are scarcely less beautiful than those
of the cirro-cumulus. It appears to have been these
clouds that in their orderly arrangement caused Ruskin
such admiring wonder. Calling them the “flocks of
Admetus” he one day essayed to estimate the number
of cloudlets present in the six bands of cloud stretched
across the sky, and found them to be approximately fifty
thousand."

When waved patterns of these two cloud-types are
seen they are generally called “mackerel skies” in
popular terminology, on account of their resemblance in
shape to the markings on the mackerel; and _ the
globular or flaked types are sometimes spoken of as
“dappled skies”.

False Cirrus.—Alto-cumulus may develop from, or
succeed cirro-cumulus when condensation is proceeding
at successively lower levels, just as alto-stratus follows
cirro-stratus, and both these alto-clouds may also be
the transformation product of another cloud which will
again be referred to when dealing with thunder-clouds
and shower-clouds. This cloud is known as “false
cirrus’ on account of its resemblance to the true or
ordinary cirrus. It is not found at the same high level
as the true cirrus, being actually only about half as high
on anaverage, but there is reason to believe that it does
not differ in actual constitution from the normal type.
Captain C. K. M. Douglas, R.A.F., has observed false
cirrus from his aeroplane, and considers it to be composed
of ice-crystals ;* Captain C. J. P. Cave, R.E., has ob-

"Ruskin, “‘ Modern Painters,” Vol. V, Part VII, chap. ii.

* Note by Sir Napier Shaw in “ Cloud-Forms according to the
International Classification,” M.O., 233, p. ro.

CLOUD-FORMS AND TRANSFORMATIONS 27

served in it portions of halos and brilliant mock suns ;!
and at Aberdeen, parhelia, paraselene, and sun and
moon pillars accompanying false cirrus have been
recorded on numerous occasions by the writer. It may
be accepted then that the false cirrus is really an ice-
crystal cloud, yet on almost every occasion that it has
been seen at Aberdeen during the last few years it has
eventually become either alto-cumulus or alto-stratus,
rather more frequently the former. A very interesting
case where the reverse process took place occurred on a
day in February, 1919. A sheet of slightly fused but
otherwise normal alto-cumulus was moving slowly from
S.S.E., and, as it approached, streams of rain or more
probably of snow were seen falling from the advancing
edge all along its front, while the alto-cumulus itself
gradually dissipated. The process continued gradually
throughout the whole sheet, with the result that in about
an hour the alto-cumulus had disappeared entirely. But
the streams of snow, which had fallen vertically at first,
seemed slowly to trail out till they took up a position
nearly horizontal, and assumed the usual appearance of
false cirrus, remaining thus for at least two hours. In
this instance the false cirrus was formed at a level some-
what below that of the original alto-cumulus.

There is no very easy rule whereby false cirrus when
seen by itself may be distinguished from the true cirrus,
nor is it actually important to do so, save as a rough in-
dication of its height, for, as Sir Napier Shaw says, “if
an ice-crystal composition be the properly distinctive
characteristic of the thread-like structure of cirrus, it only
hampers our conception of the atmospheric processes if

1 Captain C. J. P. Cave, “The Forms of Clouds,” “Q.J.R.
Met. Soc.,’”’ January, 1917, p. 73.

28 CLOUDS

we assume all clouds which show that structure to be at
a high level”! It would be better perhaps to regard the
existence of the cirrus structure as possible through a
great range of altitude, and to use the qualification of
“false” merely as an indication that the cloud is known
or believed to be at an unusually low level. A skilled
observer may frequently be able to note several small
differences between the appéarances of the high true cir-
rus and of the lower “false” type. The latter is usually
denser, or more solid in its appearance ; the fibres are
not so sharp and clear but somewhat more “ woolly ”
in their texture; the colour of the cloud is frequently
cream instead of the pure white of high cirrus; and_ the
cloud is scattered more irregularly over the sky, the
banded structure and long radiating lines, which are so
characteristic of true cirrus, being very seldom seen in
“false” cirrus (Plate 23 a).

Turret-cloud.—A very rare but very important
variety of alto-cumulus is met with in thundery weather
in summer time. Though it has not been defined and
pictured in the International Classification, yet special
mention has been made of it by Clement Ley, Clayden,
and Cave in their writings, and it is referred to as an
additional type in the small cloud atlas issued by the
Meteorological Office during the war.” The name given
to it by Clement Ley was a@lto-cumulus-castellatus, and
it is called ‘“‘turret-cloud” in English on account of its
rather unusual appearance, the cloudlets being developed
more in the vertical direction than in the horizontal, and
closely simulating, though on a very greatly reduced

1Sir Napier Shaw, M.O., 233, p. 10.
*“Cloud-Forms according to the International Classification,”
MO .5333;

CLOUD-FORMS AND TRANSFORMATIONS 29

scale, the thunder-clouds that so frequently follow their
appearance, ‘The turret-cloud is rather sporadic in its
occurrence, being seen sometimes in large quantity on
several consecutive days and then being absent from the
sky for months together. It was unusually plentiful at
Aberdeen during the summer months of 1917, particu-
larly during the thundery weather from 13th to 17th July.
All through this period the atmosphere was very much
stratified, clouds being visible at almost all levels simul-
taneously, an indication of the very disturbed conditions
which resulted in four thunderstorms in the four days.
On the 14th, following an early morning thunderstorm
characterised by very frequent and vivid lightning, the
alto-cumulus-castellatus was very finely developed.
Several patches of a very hazy purplish-grey stratiform
cloud rapidly formed, and these appeared to have a
saucer-like shape, being slightly depressed in the centre.
From out of these small sheets there rose up some
miniature clouds exactly resembling a large thunder-
cloud in shape and general appearance, but they did not
last long; rapid changes of form were taking place in
them, and they dispersed almost as rapidly as they
formed. There seems no reason why they should not
be regarded as indicating, though at a very high level,
and on a smaller scale, rapidly rising currents of warm
air similar to those which form the ordinary cumulo-
nimbus clouds that cause thunderstorms, and the im-
portance of recognising this form of alto-cumulus as
a special type lies in the fact that it is so frequently
followed by weather of an unsettled thundery type.
Lenticular Clouds.—The cirro-cumulus and_alto-
cumulus clouds are additionally noteworthy because they
appear, in certain circumstances, in another form deserv-

30 CLOUDS

ing special mention. In the International Classification
one of the notes following the descriptions of the clouds
made reference to the “ ovoid form” which occurred during
sirocco and féhn winds, and gave the word lenticularis
as its qualifying designation. But beyond mentioning
the form, no further description was given, though in
reality these lenticular clouds—or rather cloud-banks—
differ from all other cloud-formations in one important
particular ; they are not clouds in the ordinary sense of
the term, they are really places in the atmosphere where
clouds form and exist for a very short time only. They
have been termed “lenticular” from the resemblance of
their form to that of the cross-section of a lens. It is
only occasionally that a perfectly typical form is seen ;
more often several cloud-banks are joined together in
an irregular mass, but the lenticular or almond-shaped
structure is not difficult to recognise even in heavy
banks or sheets of the cloud. Their outline has been
likened to many forms, to that of an airship, to the shape
of the moving sand-dunes in the desert, to tidal sand-
banks, and to the wind-blown snow-sastrugi met with in
the polar regions and on mountain slopes; all of which
forms are of the ‘‘ stream-line”’ order.

There are occasions, though they are infrequent,
when, instead of lenticular cloud-banks being seen
against the blue sky, there will be a continuous sheet
of cloud with interstices of a lenticular shape through
which the sky is visible, but this reversed order of things
occurs in circumstances that are similar in every way
to those accompanying the ordinary lenticular clouds.
Then again, a combination of both arrangements may
be seen, where the central part of a lenticular mass is,
as it were, eaten out, thus leaving a resulting cloud-form

CLOUD-FORMS AND TRANSFORMATIONS 31

that resembles—to quote Sir Napier Shaw '—‘‘a large
horseshoe as seen from beneath at a great distance”,
The edges of these cloud-banks are usually rather ill-
defined and broken, but the windward edge is sharper
than the leeward one, and sometimes may even present
a clean-cut outline. In Fig. 5, which gives a sketch of
some typical clouds of the lenticular form, the arrow is
intended to represent the direction of the wind at the

Fic. 5.—The formation of lenticular cloud-banks.

cloud level, and consequently also the direction of
movement of the cloudlets which compose the cloud-
bank. If one of these lenticular masses be carefully
observed, it will be noticed that at the windward edge
(2) small cloud-flakes are rapidly forming, while at the
leeward edge (4) other cloud flakes are passing out from
the main mass and evaporating. Should the cloud-bank
be sufficiently thin, and chance to be so placed as to

1Sir Napier Shaw, note in M.O., 233, p. 4.

32 CLOUDS

have the sun or moon behind it, it will be quite possible
to watch the cloudlet form at the windward edge, pass
right through the cloud-bank and out at the leeward
edge, where it rapidly evaporates. Even though the
cloudlets themselves may be moving with a high velocity
—which as a matter of fact they usually do—yet the
cloud-bank may remain practically stationary for a very
long time. The cloud-banks shown in Plate 214 re-
mained in the same part of the sky for over two hours,
though the details of their structure were changing
rapidly all the time, and on another occasion, a long but
quite narrow lenticular cloud was observed to remain
stationary for a period of more than three hours, while
its component cloudlets were moving rapidly through
the mass.

Clouds having the characteristic lenticular form seem
to depend partly upon the locality where they are ob-
served, the main local influence being evidently the
presence of mountains. The winds referred to in the
International Classification, the wzzstral and’ fohn, are
both associated with mountains or plateaux. At Aber-
deen lenticular cloud-forms are seen at their best during
south-westerly to westerly winds, and it is noteworthy
that important masses of the Grampians lie to the south-
westward and westward of the city ; while Captain Cave,’
observing the same type of clouds in the south of
Hampshire, notes that they are almost exclusively con-
nected with north-westerly winds which come over the
South Downs.

An explanation of the cause of these clouds will be
given in the next chapter, but meantime a few more

‘Captain C. J. P. Cave, “The Forms of Clouds,” "ORs
Met. Soc.,” January, 1917, p. 66.

CLOUD-FORMS AND TRANSFORMATIONS _ 33

characteristics of the lenticular cloud-banks will be de-
scribed. At Aberdeen, their presence is associated on
almost every occasion with a wind which is strong or
high and of a very gusty character, and which shows
also a periodic rise and fall in its average velocity. The
temperature of the air is very much above the average
for the time of year, both in summer and in winter, when
these clouds are present, and the barometric pressure is
very unsteady, rising and falling alternately at short
intervals of time, very much as the velocity of the wind
does. In summer the colour of the sky, seen between
the clouds, is of a very intense blue ; in winter this is not
so noticeable, probably on account of the greater amount
of surface mist and smoke haze, but at all times of the
year the sunrises and sunsets are characterised by an in-
tensity of colouring which is much greater when the sky
is covered with lenticular clouds than when the ordinary
forms only are present.

On some occasions the lenticular character may be
shown only by the cloud at one particular level, but it
is far more usual for the sky to show sheets of these
clouds at several levels simultaneously, all of them ex-
hibiting the same characteristics of practical immobility
of the cloud-bank combined with rapid internal change.

There are times when the lower clouds, such as
cumulus and stratus, are seen to assume a somewhat
similar lenticular form in guze¢ weather, but in such
cases the conditions mentioned above are absent. In
some parts of the country this form of stratus is very
common and has been popularly termed the “‘ fall cloud ”,

A lto-stratus —Alto-stratus, to which reference has
already been made when speaking of cirro-stratus, is,
when seen in its most typical form, a sheet without any

3

34 CLOUDS

visible structure, and of a density just sufficient to allow
the sun or moon to be seen through it merely as a bright
blur, an appearance which has earned for it the popular
name of ‘“ watery sky,” though possibly this name has
had its origin rather more in the practically certain rain
which is the sequel to the appearance of alto-stratus.
When thin, the alto-stratus is sometimes of a yellowish-
grey colour instead of the normal bluish-grey, but when
very dense it has a dull leaden hue. In addition to
being a subsequent development of cirro-stratus, it may
also be formed by the fusion of masses of false cirrus,
while a third source of origin is to be found in the heavy
sheets of cirro-cumulus which frequently move from the
westward and which sometimes lose their normal appear-
ance by becoming fused into a continuous and almost
uniform sheet. In these last two cases, the alto-stratus
is not often structureless ; careful observation may usually
discover in it a slightly ribbed or reticulated pattern,
and occasionally its under-surface is very finely rippled,
or exhibits the appearance known as ‘ mammillation,”
that is to say, that small dome-like or partly. globular
pendent masses are hanging, very much like inverted
small cumulus, from the under-surface of the cloud-sheet.

Strato-cumulus.—Strato-cumulus, which is in reality
only a much heavier and lower variety of alto-cumulus,
shows all the main characteristics of the latter type, but
is much more irregular in its form. Huge waves or
bands of strato-cumulus are sometimes met with which
must be many miles in length, and which may be sepa-
rated by intervals of one or two miles (Plate 27 8) and
yet exhibit a wonderfully parallel arrangement in the
bands; but much more commonly the strato-cumulus
appears as a vast sheet of heavy cloud broken up into

CLOUD-FORMS AND TRANSFORMATIONS _ 3s

isolated masses by an irregular system of lanes
(Plate 258). At times the waves or bands appear to
be composed of one solid mass of condensation, while at
others they seem to be built up by the massing together
of a large number of smaller flakes (Plate 264), this
latter condition being especially well marked when the
strato-cumulus assumes the lenticular form. The edge
of a sheet of strato-cumulus may occasionally present
an appearance resembling the vertical face of a cliff, and
Captain Douglas, R.A.F., observing by aeroplane, has
met with this ‘“cliff-front” structure even within the
cloud-sheet. In one case he relates that the upper sur-
face of the cloud-sheet rose up in an almost sheer step
of 709 feet, and that the appearance when seen from
above was like that of a tidal bore flowing up a river
estuary.

Strato-cumulus is sometimes a subsequent develop-
ment of cumulus; it has been noticed frequently at
Aberdeen that towards evening the cumulus clouds, in-
stead of dispersing as is usually the case, gradually lose
their characteristic rounded summit and level base, and
flatten out into the usual shape shown by strato-cumulus
(Plate 308). The reverse process has also been ob-.
served, though not so often, and when it did occur, it
was generally during a spell when the cloud character
changed from cumulus during the day to strato-cumulus
at night, becoming cumulus again during the following
morning.

Nimbus.—It is doubtful whether Nimbus, the next
type of cloud mentioned in the International Classifica-
tion, should really be considered as a type at all. Its
essential feature—steadily falling rain or saow—is found
to accompany several cloud-sheets of quite different

36 CLOUDS

types. The rain-cloud of a cyclone, the heavy thick
layer of formless cloud with steady rain, is certainly a
form of cloud that could not be assigned to any of the
other groups mentioned in the above classification, but
then the rain may cease without any other alteration
taking place in the cloud-layer for often a very long time.
On the other hand, steady rain often falls from clouds
that show distinctly the cumulus form, and more or less
continuous rain is frequently produced by the degraded
remains of cumulo-nimbus. The matter is further com-
plicated by the fact that the ‘“‘nimbus” may consist of
more than one cloud-layer, as was demonstrated on
several occasions during the war when the searchlights
were in operation at night. Oneevening during a slight
but steady drizzle the searchlight at Aberdeen revealed
the conditions that are shown in the frontispiece. The
beam first pierced an exceedingly thin layer of broken
‘““scud”” cloud- which did not suffice to dim its brilliance,
and then entered into a layer of dim nebulous condensa-
tion which seemed rather denser in the middle and faded
into invisibility both upward and downward. Leaving
this layer the beam became almost invisible till it im-
pinged upon the under-surface of a very dense cloud-
layer, lighting it up with great brilliancy. The highest
layer was probably between one and two thousand feet
high, while the misty stratum was about half that height
and possibly three or four hundred feet thick. It would
appear that the fine drizzle was falling from the misty
layer and rendered the searchlight beam brightly visible
up to that level, while above it there was no precipita-
tion to be illuminated, and thus the beam became much
less bright.

Rain or snow showers have often been seen to fall

CLOUD-FORMS AND TRANSFORMATIONS 37

from the intermediate clouds of the alto-cumulus and
heavy cirro-cumulus types, though such showers evapor-
ate again before reaching the ground, and very fre-
quently there forms below strato-cumulus a thin film of
cloud which disperses again after a short interval by
falling in a shower of rain. Thus it appears that nim-
bus may be used as a term embracing many rain-pro-
ducing clouds of widely differing oon or arrangement,
* and Clayden,* who has commented upon the matter
in like manner, has suggested the use of the term
“nimbus ’’ as a qualifying word rather than as a type.
Cumulus and Cumulo-nimbus.—The clouds of di-
urnal ascending currents are easily recognisable on
account of their great size and massive structure. The
dome-shaped heap with a flat base is characteristic of
both cumulus and cumulo-nimbus, but the disparity in
size between them is very great. The ordinary cumulus
of a summer day is only some fifteen hundred feet thick,
but cumulo-nimbus clouds continue their growth till
sometimes they are four miles deep from summit to base,
and may spread over many square miles of country,
Their enormous size accounts for the fact that they are
best seen when distant, for as they approach, their ad-
vancing bases (which, on account of their greater prox-
imity to the observer, and in accordance with the laws of
perspective, move at a much greater apparent velocity
than do their summits) project forward and hide from
view the upper portions of the cloud. The cloud then
passes overhead as an intensely blue-black or dull purple
sheet, and after the passing of the base, the main cloud
mass again becomes visible gradually as it recedes.
Cumulus and cumulo-nimbus clouds are both the

1A. W. Clayden, “Cloud Studies,” p. 74.

38 CLOUDS

result of the condensation of moisture contained in the
rising air-currents which are caused by local heating of
the atmospheric layers in contact with theground. The
actual process will be fully described in the next chapter,
but it may be mentioned here that the cumulus clouds
are formed by the smaller “convection” currents, as
they are termed, while the cumulo-nimbus are the pro-
duction of much more vigorous ascending currents.
The flat base of the cloud marks the normal level at »
which condensation takes place in any particular instance,
while the rounded dome and columnar protuberances
mark the continued ascent of the convection currents.
That these currents may be very powerful indeed is
shown by the hailstones that accompany thunder-clouds.
Large hailstones when examined show by their internal
structure that they-are built up of successive layers of
ice, which would seem to indicate that the hailstones
have been carried up to the upper parts of the cloud
several times, receiving each time an additional coating
of ice by the freezing of the water they have collected
in the lower parts of the cloud. To support the weight
of these large hailstones and to carry them up to the
great heights sometimes reached by the top portions of
cumulo-nimbus, an upward current exceeding twenty
miles per hour has been mentioned by Captain Cave! as
being necessary. The downward air-currents are also
often very considerable, and cause great turbulence in
the cumulo-nimbus clouds, as may often be witnessed
even beneath the cloud-base, where small fragments of
cloud may be whirled rapidly downwards and upwards
by these currents. Sometimes the downward current,

1C. J. P. Cave, “The Forms of Clouds,” “Q.J.R. Met. Soc,”
p. 74.

CLOUD-FORMS AND TRANSFORMATIONS 39

especially if accompanied by hail, is seen to carry down
with it parts of the base of the cloud, and to cause them
to hang in huge bulges below the main mass. This
usually occurs in the rear portion of a thunder-cloud or
of a hail-squall cloud, and the appearance is termed
manimato-cumulus,

One special feature of cumulo-nimbus is the develop-
ment above the domed portion of a mass of condensa-
tion which has a shape very closely resembling an anvil,
if judged by its profile appearance, but which is really a
tabular flat-topped mass, rounded in plan and widest at
the top. The edges of this “anvil” are frayed out into
the fibrous form associated with the cirrus clouds, and
on this account the anvils are said to form the “false”
cirrus which has already been described in this chapter.
Clouds showing the anvil form (Plates 31 & 32) are usually
regarded as the type which gives rise to thunderstorms
with hail and rain. This is undoubtedly true, but it is
worthy of mention that two severe though short thunder-
storms which occurred at Aberdeen within recent years,
and which were accompanied by heavy rain and _ hail,
proceeded from clouds which retained their hard rounded
outlines without any sign of the anvil form developing.
On the other hand, in spring and autumn, a much smaller
though still extensive form of cumulo-nimbus is very
frequent at Aberdeen, and always shows the anvil
structure exceedingly well developed, so much so in
some cases that the cloud appears to be almost entirely
anvil (Plate 324). This particular form is always ac-
companied by wild squally weather with frequent showers
of rain, sleet, and hail.

The gradual growth of the anvil form is shown in
the diagram Fig. 6, where there are shown the changes

he CLOUDS

Fic. 6.—Growth and development of a cumulo-nimbus *

anvil’’,

CLOUD-FORMS AND TRANSFORMATIONS 41

that took place during approximately half an hour in a
large cloud, very far distant beyond the horizon, one day
in the autumn of 1918. The cloud was moving from
west-north-west, that is, away from and a little to
the right of the observer, who was looking due east.
The anvil will be seen apparently moving slowly to the
left, which indicates in reality that the air-currents at the
level of its top were more westerly than were the surface
ones, and the last phase seems to show that the ascend-
ing currents were losing their vigour, since the main
mass of the cloud appears flatter in that phase. Alto-
gether the six phases serve to demonstrate how very
great may be the changes that can take place in a short
space of time, even in the largest clouds.

When, after a period of activity which may extend
sometimes over several days, the ascensional currents
become more feeble, or cease altogether, the anvils of
false cirrus usually become separated from the cumulus
mass, and, assuming sometimes the forms of patches of
alto-stratus, alto-cumulus, or even of thin strato-cumulus,
they may drift away without evaporating, and thus it
happens that they may be seen in regions far away from
their origin. At Aberdeen these cloud patches often
herald the approach of weather similar to that which
produced them, particularly when they are moving from
the north-north-west.

If cumulo-nimbus clouds become grouped together
into a very large mass, as is very commonly the case in
thunderstorms, it may occasionally be noticed that the
upper portions of the cloud are moving from some
southerly point, while the base may be moving from
another totally different direction, generally from north-
west or north. The cause of this is probably that a

42 CLOUDS

north-westerly air-current, which is considerably cooler
than the rest of the air in the neighbourhood, undercuts
the southerly one, and in doing so forces up the warmer
and lighter surface air, thus increasing the convection
currents and consequently also the formation of cloudy
condensation. This gives rise eventually to thunder
and hail and rain squalls, and the coolness which suc-
ceeds the thunderstorm and its previous sultry conditions
has been in reality the cause and not the result of the
storm, though the process is popularly referred to as “‘ the
thunder clearing the air”. Such thunderstorms are of
the nature of the “line-squall” which will be described
later.

An outstanding feature of cumulus and cumulo-
nimbus clouds, that which segregates them in a section
by themselves, is their liability to appear in greatest
quantity at one particular time of the day. During
weather when their type is prevalent the early morning
may present a sky of cloudless blue, but somewhere be-
tween eight and ten o'clock a few small fragments of
white cloud appear and increase rather rapidly in size
till they become the familiar dome-shaped heap clouds
known as cumulus. As time goes on these clouds grow
larger and more numerous, till the sky may become
almost entirely covered with them. The period of their
maximum is usually found to be about two or three
o'clock in the afternoon. After this they gradually
diminish both in quantity and in size till towards evening,
when they may have vanished altogether, leaving the
sky once more cloudless. Sometimes the type of cloud
may remain cumulus throughout, the heaps growing”
very little higher, though much more numerous—at times
sO numerous as to coalesce and completely cover the

CLOUD-FORMS AND TRANSFORMATIONS 43

sky—but on other occasions they may be seen to
increase in the vertical direction and to become cumulo-
nimbus. This latter form also may disperse in the
evening, but ina different manner. The upper portions
of the cumulo-nimbus cloud frequently lose their cumulus
form and either flatten out into a sheet or into a number
of fragments resembling thin strato-cumulus. In so doing
they become detached from the lower parts of the cloud
and float away exactly as do the “anvils” of false cirrus,
while the basal part of the cloud disperses entirely. It
is only comparatively rarely that cumulus clouds are seen
during the night-time, though cumulo-nimbus are rather
more common, but in all such cases their presence is a
sign of atmospheric instability.

The showers which fall from cumulo-nimbus clouds
are responsible for the appearance of that most beautiful
of all optical phenomena, the rainbow. Rainbows are
seen on the opposite side of the observer from the sun,
and appear as coloured arcs low on the horizon if the
sun be high in the heavens, and becoming higher and
more extensive as the sun declines, till they are seen at
their greatest extent as semicircles when the sun is just
setting. An imaginary line from the sun's disc passing
through the head of the observer will, if projected upon
the plane of the falling shower, give the centre of the
circle of which the rainbow forms an arc. As the
angular radius of the primary bow is approximately 42°
it follows that no such rainbow can be seen if the sun's
angular elevation above the horizon is more than 42°.
Occasionally there is seen a “double rainbow,” that is,
a secondary bow outside the primary one, the angular
radius of this secondary bow being approximately 54°.
The secondary bow is fainter than the primary and has

44 CLOUDS

the order of the colours reversed. The primary bow
shows in ordinary circumstances the complete spectrum
from red to violet, the red being on the outside of the -
circle, followed by orange, yellow, green, blue, and violet
in the order mentioned. In the secondary bow the
violet is outside and the red inside.

The bows are caused by the refraction and reflection
of the sun’s rays in the drops of rain. To form the
primary bow the ray is first refracted, then reflected once
within the raindrop, then again refracted, while in the
case of the secondary bow, the first refraction is followed
by two reflections before the second refraction occurs.
Within the primary bow there are often seen additional
narrow bands which immediately follow the violet, and
which are known as supernumerary bows; the usual
colours of these bands are alternately a faint magenta
and dull blue-green. ‘They are caused by the diffraction
of light reflected from the drops. The rays which do
not follow the path of minimum deviation are also re-
flected, and produce immediately within the primary bow
a general brightnéss which contrasts very markedly at
times with the darker area outside the bow. This dif-
ference of tone as well as portions of the additional bands
above-mentioned are shown in Plate 35 ns.

Originally it was supposed that all rainbows were
of the same size, and showed the same colours, but
closer observation has proved that slight variations occur
in the size of the bow, in the width of the colour bands,
in the relative intensity of the various colours, and even
in the colours themselves. On one occasion the sequence
of colours as observed by Mr. F. J. W. Whipple! were

‘Note by F. J. W. Whipple, M.A., in Meteorological Office
Circular, No. 4, Sept., 1916.

CLOUD-FORMS AND TRANSFORMATIONS 45

as follows—orange, yellow, greenish-blue, indigo, pink,
green, pink, neutral tint, pink, the colours following the
indigo being presumably the additional bands already
mentioned. The sequence of colours depends upon the
size of the raindrops which give rise to the bow.! In
very hazy weather, especially if the sun be low, it is not
uncommon for a bow to be seen having no colours other
than the red and orange visible; this is due, of course,
to the selective scattering of the green, blue, and violet
rays of the sun’s light by the haze and smoke in the
atmosphere.

Rainbows may be formed by moonlight as well as by
sunlight, but their colours are much less brilliant than in
the case of the daytime bows, being seen usually with
difficulty, and often the bow appears colourless.

Stratus —The lowest type of cloud, known as stratus,
is, as its name implies, simply a sheet of condensation,
In typical form its appearance is that of a uniform grey
pall, usually a little lighter towards the horizon and
darker overhead. Beneath it the atmosphere is some-
times excessively clear, objects like ships on the sea being
seen at a distance of ten or twelve miles as jet black
silhouettes, and as clearly defined as objects near at
hand. :

At other times there may be a general mistiness of
the air between the under-surface of the stratus cloud
and the earth. In special cases there may be no definite
line of demarcation between the mist and the cloud, and
in the extreme case the cloud itself may rest upon the
earth, when it is known asa fog. A fog which clears
from the surface of the ground by elevation becomes
a sheet of typical stratus; the process is remarkably

1 Pernter, “‘ Meteorologische Optik,” p. 540.

46 CLOUDS

common at Aberdeen during spells of sea fog, when at
one particular part of the day the town may be shrouded
in fog, while an hour or so later everything will be clear
below for some little time, with a sheet of stratus cover-
ing the sky, only to revert again to the fog condition
later on.

Stratus is sometimes an unusually persistent cloud,
spells of three to five days of unbroken cloud being not
uncommon at Aberdeen. It is during such spells that
the remarkable visibility above-mentioned is met with.
When stratus sheets break up and disperse they often
do so by forming into lumpy rounded masses very much
resembling rather flattened cumulus. This is the form
to which the International Classification alluded as de-
serving the qualifying term ‘‘cumuliformis ” to be added
to its type-name. Examples bearing a very close re-
semblance to cumulus have been seen often at Aberdeen ;
on one occasion the form was so massive and rounded
and showed so flat a base that had it not been seen to
form from the stratus it certainly would have been classed
as cumulus. Its height was very small, a pilot balloon
which was sent up entered the cloud at about eight or
nine hundred feet. On other occasions the sheet of
stratus separates into masses similar to those of strato-
cumulus, and even into long rolls and waves, but usually
these masses and waves are so thin that the sun’s disc
may be seen quite clearly through them (Plate 34 8).
Even the largest masses of dispersing stratus are re-
latively thin. Another method of dispersal is for the
sheet to break up into ragged shreds, particularly if the
sheet itself has recently been a ground fog, and at other
times the cloud separates into very thin wreaths of vapour
which, streaming along on the wind, flow in undulating

CLOUD-FORMS AND TRANSFORMATIONS 47

lines which thus delineate the actual paths of the air-
currents in which they are borne.

Detached sheets of stratus are often seen in con-
junction with other cloud types. At Aberdeen one of
the most common of these occurrences is for a layer. of
strato-cumulus to have immediately beneath it a stratus
film of greater or lesser density ; this film may at times
give some slight precipitation, as has been already
mentioned when dealing with nimbus clouds. Frag-
mentary stratus is also very common during certain types
of weather in winter time at Aberdeen; these will be
mentioned in a Jater chapter. In thundery weather at
this station part of the sky may suddenly be covered by
an indefinite purplish-grey area of gloom, which is with-
out doubt true stratus; it vanishes as rapidly as it ap-
pears, only to form again later on. Another rather more
stable form of stratus is found in what appear to be lines,
but which are really slightly lenticular patches of cloud
floating at several levels simultaneously round the flanks
of massive thunder-clouds, to which they give a very dis-
tinctive appearance.

The Colours of the Clouds.—The colours of the clouds
in daytime are for the most part white and grey, white
where the full sunlight falls upon their crests or upper
surfaces, and all the most delicate tones of grey in their
shadowed portions. In the larger clouds, the grey may
become very dark ; when it does so it usually shows a
slightly bluish tint. In thunder-clouds the white may
assume a yellowish or even coppery-red hue, and the
darker parts may become purple-grey, this is caused by
the soot and smoke or even dust that are carried up into
the clouds by the powerful ascending currents which
form them.

48 CLOUDS

But when the sun is low in the heavens, as during —
the daytime in winter, or towards the time of sunset,
the white colour of the clouds takes on a golden tinge,
which grows deeper and passes through various tones of
orange and red as the sun sets, or in the reverse order
as it rises. The dust particles and haze in the atmos-
phere cause a selective scattering of the sun’s rays,
stopping out those of shorter wave length like the violet
and blue, and permitting only the orange and red to
pass. This is done all the more effectively because the
rays of light from the setting sun are refracted by our
atmosphere and pass for a very great distance through
the lower layers of the air wherein most of the dust haze
is found.

The lower clouds are, of course, the first to assume
the ruddy hues of sunset, and it is no uncommon sight
to see a sheet of cirro-cumulus still dazzlingly white
while the strato-cumulus below is deep orange, but, as
time passes, the lower clouds gradually lose their red
colour and become deep purple-grey while the sunset
hues are transferred to the upper layers. If the cirro-
cumulus clouds are high and of delicate structure they
become suffused with the most beautiful rose-colour, a
tint far purer and more ethereal than is ever to be
seen on the lower clouds, and which, when seen against
a deep blue sky affords a picture of exquisite beauty.

But the most striking sunsets seen at Aberdeen are
those which accompany the presence of the intermediate
clouds of the lenticular form. Frequently when these
are present there is a narrow belt of cloudless horizon
in the west, so that no obstruction is offered to the last
rays of the setting sun. Then again, the lenticular clouds
themselves have most beautifully rippled under-surfaces,

CLOUD-FORMS AND TRANSFORMATIONS 49

and this greatly enhances the effect of the colour, for the
sky may at times look like one vast inverted sea of
flame, or, as Ruskin so beautifully has expressed it, be.
covered “as with the drifted wings of many companies
of angels’’.* And another peculiarity of these lenticular
skies is the very long duration of colour that sometimes
occurs, One such display—a most extraordinary one—
was seen at Aberdeen in January, 1911, when the clouds
showed varying colours for over two hours. Shortly
before four o'clock in the afternoon the edges of the
clouds were outlined in pale gold, which gradually grew
more intense until, when the colour became transferred
to the under-surfaces of the clouds, they gleamed like
polished brass. As time went on the yellow deepened
into orange, which in turn flushed into flame colour and
scarlet, the open sky beyond the clouds changing at the
same time from pure blue to greenish-blue. By five
o'clock the clouds were vivid crimson and the sky a clear
pale emerald, but the colour display was by no means
over, for the clouds continued slowly to deepen in hue
till as late as six o'clock when they showed a most
wonderful colour, one not easy to describe, anangry red,
so deep as to be almost black, yet nevertheless of great
intensity. The horizonsky had changed to a dull milky-
white hue.

When it happens that, after a spell of rainy weather,
the nimbus clouds commence to break up about the
time of sunset and to assume a somewhat globular form
resembling strato-cumulus there is occasionally to be
seen a curious play of colour in them. Separate masses
will appear suffused in turn with dull pink, violet, dull

1 Ruskin, ‘Modern Painters,” Vol. I, Part II, section 11,

chap. iv.
4

Po) CLOUDS

orange, and even green. The colours are additional to
their own greyish hues, and are therefore subdued but
are none the less very evident.

Another variety of cloud colouring, which is seen in
full daylight, is that known as “‘irisation”. The pheno-
menon is seen best developed in the higher clouds of the
alto-cumulus and cirro-cumulus types, particularly in that
form of the latter which is found massed in the oval
lenticular sheets having a slightly rippled structure.
The colouring is seen at angular distances from the sun
varying mostly between to’ and 30°, and also shows
a strong tendency to appear along a horizontal line
through the sun’s disc, though it may also appear other-
where. The colours shown usually follow the edges or
contours of the cloud masses, and are not pure spectrum
colours but compounds of these. The commonest are
rose-pink and emerald-green, both of which are usually
very pure and bright, while yellowish-orange, violet-blue,
and greenish-yellow are also more or less in evidence.
In fact the iridescence resembles very closely the appear-
ance of mother-of-pearl, and Dr. Simpson * has explained
it as being due to diffraction effects produced by the
various sizes of water-droplets present in the cloud.
Dr. Simpson, who accompanied Scott’s last expedition,
made a special study of the iridescent clouds which were
so frequently seen in the Antarctic.

1G. C. Simpson, D.Sc., ‘“Coronz and Iridescent Clouds,”
*Q.J.R. Met. Soc.,” October, 1912.

CEUAP LE ReLV;
SOME CAUSES OF THE FORMATION OF CLOUDS.

Couns are really the visible products of the condensa-
tion of the water-vapour that is always present in the
atmosphere. ‘The heat of the sun’s rays is constantly
evaporating moisture from the surfaces of the sea and
the land, and the invisible water-vapour thus produced
is absorbed by and diffused in the air. The air will
continue to take up water-vapour and the latter will re-
main quite invisible until a point is reached when the air
becomes saturated, and then any further addition of
water-vapour results in super-saturation, and the excess
of moisture becomes visible as a cloud, provided that
there are present in the air some nuclei upon which the
condensation can form. As the lower layers of the air
always contain a good deal of small dust particles in
suspension, the required nuclei are always present there,
and it has been demonstrated that ionised air itself may
also act in the same way as material nuclei do, so that
condensation thus becomes possible wherever water-
vapour may exist in sufficient quantity. This sufficiency
of quantity varies very greatly. The amount of water-
vapour that a given mass of air can contain before
becoming super-saturated depends entirely upon the
temperature of the air, no other condition affecting it in
any way. ‘The surface layers of the air are very much
warmer than those above them, as will be explained

cas oe,

52 CEOUDS

later, and the capacity of the air for water-vapour in-
creases at a much greater rate than does the temperature
necessary to produce this capacity. At a height of 5
kilometres (or 3 miles) a given volume of air can con-
tain less than one-fifth of the water-vapour that it can
hold at the surface, while at double that height the
amount of water-vapour present must be less than one-
thirtieth of the surface quantity. It follows, therefore,
that cloud will be found in greatest quantity usually in
the lower layers of the atmosphere, and also that the
lower clouds will often be of very great thickness while
the higher clouds will, on the contrary, be very tenuous.
If then the air in any part of the atmosphere is
saturated, and the temperature falls, the excess of mois-
ture immediately becomes visible as a cloud of water-
drops or of ice-crystals. These droplets are of exceed-
ingly minute dimensions, and therefore fall so slowly
that even the very feeble circulating air currents within
the cloud are sufficient to keep them suspended, and it
is not until increasing condensation and the consequent
coalescence or enlargement of the drops takes place that
the latter fall as rain. But though in calm air the drop-
lets of water fall very slowly, yet they do eventually
reach a lower altitude, and there they usually find them-
selves in either a warmer layer of air, or a less humid
one, wherein they are soon evaporated. But meantime
fresh drops are condensing in the cloud layer and these
cause the cloud to persist as a whole, sometimes for a
long period, though its details may vary considerably on
account of the incessant condensation and evaporation
going on within it.
Modes of Cooling of the Air.—The requisite cooling

of the air to produce condensation may be brought

CAUSES OF THE FORMATION OF CLOUDS 53

about in several ways : by dynamical cooling, where the
rising of the moist air results in its expansion and the
consequent lowering of its temperature ; by the turbu-
lence of warm moist air flowing over much cooler sur-
faces ; by a local or general fall of barometric pressure ;
by the mixing of two masses of saturated air of different
temperatures ; by the radiation of heat from the layers
of air ; and by the conduction of heat from one air-layer
to another. Of these processes the first five are the
more important, the last one is practically negligible.

1. By the Rising of the Air.—There are two methods
whereby the moist air may be transferred to a higher
level. The first is the forced rising of the warm mois-
ture-laden southerly current by the cooler air currents
from the east and north, which all combine to form
the circulation found in low-pressure areas. When this
takes place, extensive layers of cloud are formed, which
eventually become a sheet of nimbus from which the
rainfall of the depression is produced.* Such cloud
usually occurs in a widespread unbroken sheet, which
towards its rear edge separates into shreds or else as-
sumes to some extent the form produced by the second
of the two possible methods.

The movements of these air-currents depend upon
the distribution of pressure at the time. If the barom-
eters at a large number of stations over a wide area are
read simultaneously, and the corrected readings are
plotted on a chart, then it will be found that, if those
stations where readings are practically the same be
joined by lines, these lines, which are termed zsodars
(ie. lines of equal pressure), will exhibit approximately

1 The process was very clearly explained in a lecture by Prof. V.
Bjerknes at the Royal Meteorological Society on 7th November, 1919.

54 CLOUDS

concentric and roughly circular or oval arrangements.
At times the central region will show the highest pressure
readings, and the distribution is then known as an anto-
cyclone, while, when the central area shows the lowest
pressures, it is called a cyclone. Some of the commoner
types of pressure distribution are shown in Figs. 16 and
17 in Chapter VI. It is convenient to imagine the anti-
cyclone as a hill and the cyclone as a hollow, because by
so doing it becomes possible to imagine a slope from the
one to the other. This slope is in reality one of pressure,
and the rate of slope is known as the barometric gradient.
It is of great importance in weather production, firstly
because the steeper the gradient, or in other words, the
closer the isobaric lines representing certain definite and
regular intervals of pressure are crowded, the more violent
will be the wind, and secondly, because the wind direc-
tion conforms roughly to that of the run of the isobars.
In the northern hemisphere the main flow of the surface
air in our latitudes is from the south-westward, and it is
found that the wind-currents round a low-pressure area
move in a counter-clockwise direction, while round a
high they travel clockwise. But there is a slight differ-
ence between the wind direction and the run of the iso-
bar in both cases. The surface winds round a cyclone
are inclined inwards at an angle of about 30° to the tan-
gent to the isobar, while round the anticyclone there is
an outward inclination of about the same amount. In
the upper regions, where cirrus cloud is found, this ar-
rangement is reversed, the winds of the cyclone moving
outwards, and those of the anticyclone moving inwards
(Fig. 7). There is thus shown a movement of air from
the high towards the low at the surface, while in the
upper air the motion is from low to high, or it may per-

CAUSES OF THE FORMATION OF CLOUDS 55

haps be better expressed by saying that the surface air-
currents feeding a low-pressure area are convergent,
while those associated with a high are divergent,

It is not, however, the same body of air that whirls
round a depression in a circle. In actuality masses of
air are drawn from different sources round the low pres-
sure area, enter into the system along certain paths and:
then either die out in the centre by rising, or else pass
out of the disturbance again. Shaw and Lempfert!
have investigated a large number of depressions, and
have succeeded in tracing the paths or trajectories of
the moving air involved in the disturbances, and the
same has been done in the case of the storm of 11th to
13th November, 1915, by Geddes.? From the maps of
trajectories worked out by Dr. Geddes, Fig. 8 has been
drawn. It shows in heavy lines the paths of moving
air from 4 p.m. till midnight on the 11th, while in light
lines are two other paths from 7 a.m. onwards on the
12th. The crosses on the lines indicate intervals of
two hours. The warm south-westerly current which
was at a temperature of between 50 and 60° F. may
be seen apparently to end in the centre of England, and
was evidently forced to rise over the barrier formed by
the cold easterly current which had a temperature of
from 35° to 45. F. The two trajectories from north-
ward in the rear of the depression may also have origi-
nated somewhere to the eastward. During the passage

1W. N. Shaw, Sc.D., F.R.S., and R. G. K. Lempfert, M.A.,
“Life History of Surface Air Currents,” M.O., 174, 1906.

2Lieut. A. E. M. Geddes, R.E., ‘The Storm of rith to 13th
November, 1915, in its Passage over the British Isles,” $200. F.dh.
Met. Soc.,” January, 1917.

rks CLOUDS

of the storm the British Isles were practically covered
with nimbus cloud which yielded very heavy rain.

The second form of dynamical cooling is that com-
monly known by the name of “convection,” and the
agency which produces it is. the heating of the surface
air-layers combined with the effect of the rate of fall of
temperature with height. The forms of cloud produced

Fic. 7.—Directions of air-motion round low and high-pressure systems. The
full arrows show the directions at the surface, while the dotted arrows
indicate those at high levels.

by this process are of the cumulus and cumulo-nimbus
types, and also include the long extended bank of cloud
usually associated with line-squalls. :

It is a fact of observation that, taken on the whole,
the surface layers of the air are the warmest, and that,
as we ascend, the temperature decreases in the layers
above, up to a height of about seven miles. If it were
possible thoroughly to mix up the whole atmosphere
until each layer possessed the appropriate temperature

CAUSES OF THE FORMATION OF CLOUDS 57

corresponding to its pressure, it would be found that, pro-
vided no part of the air was saturated with water-vapour,

Fic. 8.—Map showing the trajectories of air on 11th November, 1915.

the temperature of the air would decrease steadily at a
rate of about 10 C. for every 1000 metres of ascent
or about 54° F. for every 1000 feet. This rate is known
as the adiabatic lapse-rate for dry air. It implies that

58 - CLOUDS

if a mass of air at the surface were raised upward
without having any other heat communicated to it or
withdrawn from it, its temperature would fall at the
above-mentioned rate simply on account of its expan-
sion.

But this ideal state of matters does not exist through-
out our atmosphere. Irregular heating, condensation of
saturated air, convection, radiation and conduction of
heat all tend to produce something differing from adia-
batic rate. It is not intended in the present work to do
more than to indicate the salient facts, as the subject
is thoroughly dealt with by Shaw, Gold, Dines, and
others.’ In the actual atmosphere the average fall of
temperature with height is found, from the records of
registering balloons, to be about 6° C. per kilometre or
about 3 F. for every tooo feet, but the decrease is not
regular. For the first 3 kilometres or 10,000 feet the
temperature changes are erratic; after a steady fall there
will sometimes appear an abrupt rise, known as an
tnversion, which may be as slight as 1 F. and at other
times may be as much as 10 F. This rise of tempera-
ture is, however, only local and does not extend far ; the
temperature above it continues to fall off at an increasing
rate till at about 7 kilometres or 23,000 feet the rate
of fall is greatest, sometimes approaching the adiabatic
rate; but when 11 kilometres or 7 miles is reached the
fall of temperature abruptly ceases, and the tempera-
ture remains practically constant from that point upwards

‘Sir Napier Shaw, F.R.S., ‘‘ Manual of Meteorology,” Part IV,
tg1g. E. Gold, M.A., “The International Kite and Balloon
Ascents,” ‘‘Geophysical Memoirs,” No. 5, M.O., 210 (e), 1913.
W. H. Dines, F.R.S., “Characteristics of the Free Atmosphere,”
“‘Geophysical Memoirs,” No. 13, M.O., 220 (c), 1919.

CAUSES OF THE FORMATION OF CLOUDS 59

as far as balloons have as yet penetrated, which distance
is about twice the last-mentioned height. There exists
therefore an isothermal layer in the atmosphere at least
seven miles deep, superposed upon a layer about seven
miles deep wherein changes of temperature with height
are very great. On account of the great difference in
their physical conditions the isothermal layer has been
named the s¢vatosphere, while the lower layer is known

-200 -I60 -120 -80 ~40 0 40 80°F

14

(2
7)
£10
&G 4
£ AS
28 =
< c
<6 =
g ; =
S s
22

40. 160 180 200 220 240 «260 «280 300°
Absolute

Fic. 9.—Adiabatic lapse-rate (straight line) and an example of vertical tempera-
ture gradient in the atmosphere (irregular line). The base of the strato-
sphere is shown by the dotted line.

as the ¢vofosphere. The actual conditions in the atmos-

phere are shown in Fig. 9, wherein the adiabatic lapse-

rate appears as a straight line, and the average vertical:
temperature gradient as an irregular line, showing two
of the smaller abrupt “inversions”. The height of the
commencement of the stratosphere as mentioned and
shown here refers to our own latitudes ; the height is

greater (17 kilometres) at the equator, and lower (8

kilometres) at the poles. This height is important, in-

asmuch as the cessation of the fall of temperature imposes

re CLOUDS

a limit upon the possibility of convection in the strato-
sphere, and therefore clouds are not likely to form there.

In the.troposphere, however, the reverse is the case ;
convection is often very active in the surface layers of
the air, and results in the piling up of huge masses of
condensation. During bright sunny weather the ground
becomes much warmer in some places than in others,
and communicates its heat to the air layers in contact
with it. These air masses, becoming thus warmer than

5000

Fic, 10.—Diagram showing lapse-rates and condensation,

the surrounding masses, also become lighter and are
therefore forced to ascend. As they ascend they expand
and cool down at the adiabatic rate, but their ascent con-
tinues until they reach a level where the surrounding air
is at the same temperature as that to which they have
cooled, after which any further ascent would make them
colder and therefore heavier than the surrounding air.
How soon this may occur depends upon two factors ; the
actual lapse-rate in the atmosphere, and the degree of
saturation of the rising air. Fig. 10 shows a concrete ex-

CAUSES:OF THE FORMATION OF CLOUDS 61

ample. The actual lapse-rate is assumed to be 3° F. for
1000 feet and is shown by the line L. The mass of air
is taken to be 5° F. warmer than the surrounding air,
which latter is at a temperature of 55° F. As the heated
air rises, it will follow the line A, which indicates the
adiabatic rate for dry air. The two lines meet at the
point H and this indicates that the air will rise to 2000
feet and that its temperature will then be 49° F. That
is, if the air be dry or at least still unsaturated. But,
if the air contained much moisture originally, it will reach
a point during the course of its ascent where it will be-
come saturated and the moisture will condense and form
a cloud. Suppose that the relative humidity of the air
when it leaves the surface is 76 per cent. ; then the dew-
point for that humidity for air at 60° F. is 52°5° F., and
as soon as that temperature is reached, the air becomes
saturated and a cloud forms at the level C, where the
adiabatic lapse-line cuts the line representing 52°5° F.
From this point upwards the air continues rising, but
owing to its saturated condition it cools down more
slowly, because the heat produced by the condensation
of the water-vapour diminishes the rate of cooling ; the
actual rate at which it does cool is on an average ap-
proximately 3° F. for tooo feet. . Now this rate is the
same as the average rate of fall of temperature in the
atmosphere, and therefore if the average atmospheric
lapse-rate were perfectly regular, there would be no point
where the saturated lapse-line (marked S) would intersect
the line L, and therefore the ascending air would go on
rising indefinitely to the top of the troposphere. But the
atmospheric lapse-rate is not regular, as may be seen
from the dotted line F, which represents an actual lapse-
line recorded at South Farnborough. Suppose that at

62 CLOUDS

2500 feet the temperature of the free atmosphere rises
abruptly 3° F. and that this inversion makes the lapse-
line follow the line I. Then the rising columns of
saturated air will be definitely stopped at the level O,
because now the rising air is colder and heavier than the
air above it, and any further condensation will form asa
flattened layer at that level. It is because the saturated
air can continue to rise and to condense that the cumulus
and cumulo-nimbus clouds are so massive and so deep.
The more rapid the lapse-rate in the atmosphere the

Fic. 11,—Diagrams of air-motions in a line-squall. (a) Horizontal plan.
(6) Vertical cross-section. The cold air is indicated by c.

higher the ascending air columns will reach, while if the
rate is slow or negative (as shown in the line F), the
ascent of the columns will be stopped very early, or pre-
vented altogether. When the lapse-rate in the atmos-
phere approaches the adiabatic rate of cooling, the atmos-
phere is said to be in unstable equilibrium, while if the
rate is slow, or particularly when it shows an inversion,
the conditions are said to be stable.

There is one particular example of the production of
cloud by forced convection that deserves special atten-
tion; it is the long cloud-belt associated with a “ line-

CAUSES, OF THE FORMATION OF CLOUDS 63

squall”. It sometimes happens that when a current of
warm air from the south or south-west is flowing across
our islands it is invaded on its western side by a cold
_ west or north-west current (Fig. 11 (a)). This westerly
current is colder and denser than the southerly one, and
consequently the latter is forced up by it, somewhat as
shown in Fig. 11 (4). The forcing-up of the southerly
current causes the warm air to expand and cool, and the
moisture contained in it condenses in a band of cloud
resembling cumulus, which stretches usually in an un-
broken line right across the sky, and moves forward with
the squall front very much as does the foam on a break-
ing wave, or the tidal bore up a river estuary. Line-
squall fronts often extend over hundreds of miles and are
often accompanied by a destructive short squall of wind,
a shower, and occasionally by thunder and lightning,
The wind veers round suddenly from south to west as
the cloud passes overhead, the barometer shows a sudden
* small rise, and temperature falls suddenly and usually
considerably. The records of the meteorological ele-
ments in a typical line-squall are shown in Fig. 12.

The structure and theory of line-squalls have been
described by Lempfert and Corless,’ and also by Shaw.”
Illustrations of the cloud phenomena accompanying the
line-squall of 14th October, 1912, at Aberdeen, are shown
in Plates 36 and 37, and the following notes were made
by the writer at the time of its occurrence :—

“ Rain had been falling all the morning until 7 o'clock,

1R, G. K, Lempfert, M.A., and R. Corless, M.A., ‘ Line-Squalls
and Associated Phenomena,” “Q.J.R. Met. Soc.,” Vol. 36, pp. 135-
170.

2W.N. Shaw, F.R.S., Sc.D., ‘‘ Forecasting Weather,” Chapter
WALES, 2053.

64 CLOUDS

after which the nimbus cloud cleared away, showing a
sheet of grey alto-stratus above. Temperature was
rising and stood at 59° F. at 10.20 a.m., and a gentle to
moderate wind was blowing from S.S.W. At about
10.24 a.m, a sudden puff of wind followed by sharper

Semis

Passage of Squall.

sacha! :

SN aecinieawentonenr gs. wipes sae ole jon

re 29.6 Pressure
4.

80° 1
© 50° _— a s Temperature.
<

40° —— = \

£ 60
3 40 Wind
8 20 Velacity
= 0
N.
iS W. Wind
= s Direction
a tS
N. -
% 0.00
< 0.10 Rainfall
$2.20

’

Fic. 12.—Records of meteorological instruments during the passage ofa
line-squall,

gusts came from the W.N.W., and a patch of cloud
formed rapidly in the S.W., lying along a line from
S.S.W. to N.N.E. (Plate 36 A). This cloud rapidly
extended along its length, moving at the same time
quickly towards the E.S.E. By 10.26 am. the cloud
had become a long band of cumulus cloud stretching
right across the sky (Plate 368), and was followed in its

CAUSES OF THE FORMATION OF CLOUDS 65

rear by a sheet of dark grey cloud. The cloud-band
and sheet passed rapidly eastward, and as it reached the
coast-line the long band of cumulus gave place to a
ragged form. At 10.28 a.m. the cloud-bank had passed
out over the sea and almost suddenly a large number of
wisps of cloud formed beneath the cloud-bank (Plate 37 4)
and rose up rapidly into the main cloud mass. The
air over the sea, being moister than that over the land,
required less elevation than the latter did to produce
_ condensation when undercut by the cold current.
Following the formation of these wisps of cloud, the
whole cloud front at 10.30 a.m. appeared, as shown in
Plate 37 8, to be a mass of vapour whirling in vortices,
while three waterspouts formed over the sea, moving
downwards from the cloud but not reaching the water,
which latter, however, showed their influence by the dis-
turbance of its surface. At the same moment there came
the heavy squall of 53 miles an hour from W.N.W.,
and the temperature dropped to 48° F., a fall of 11° F.
A-sharp shower accompanied the squall. ‘To the north-
ward of Aberdeen the squall was evidently heavier, for
factory windows were blown in and whirlwinds of dust
and dead leaves occurred. After the squall had passed,
the upper cloud was seen to be moving from the south-
west, so that the influence of the W.N.W. current was
felt only in the lower layers of the atmosphere, and the
cold current was therefore not very deep.”

A pilot balloon sent up during one of these squalls
by Dr. Geddes and the writer showed by its trajectory
that the air in the rear of the squall cloud was descending
at the rate of about four miles per hour, while moving
forward simultaneously at 48 miles per hour.

But it does not always happen that the inflow of cold

5

66 CLOUDS

air is attended by violent squalls. Many cases occur
where the velocity of the cold current, though sufficient
to undercut the warm air and to cause the line of cloud,
is not sufficient to cause anything other than a slight
gust. In such cases the cloud formed is less like a
breaker, and shows simply as a line of connected cumulus
clouds. An example’is shown in the coloured Plate
4, where a bank of massive cumulus clouds is moving
slowly to seaward, with the usual condensation wisps
rising up to meet it.

It is when the atmosphere is in markedly unstable
equilibrium that the cumulo-nimbus clouds are formed.
A plentiful supply of warm moist air and no check to
the ascension of the currents permits the summits of
thunder-clouds to rise to very great altitudes, as has
already been mentioned. The apices of these clouds at
such great heights are at temperatures far below the
freezing-point, and though water may become super-
cooled to a very marked degree before freezing (Dr.
G. C. Simpson mentions water-drops at — 21° F. in the
Antarctic), yet freezing usually does occur eventually, and
when this occurs the large anvil-shaped tops are formed
to these clouds. The anvils consist of snow mixed with
ice-crystals and spread out at the edges in fringes of false
cirrus. Thunder-clouds of large size are generally built
up somewhat in the manner shown in the sectional view
in Fig. 13. In front the ascending warm currents move
in the direction shown by the arrows at A, while in the
rear the air, cooled by the falling rain and hail, descends
as indicated at B; and underneath the base of the cloud
the opposing currents set up at C the turbulent motion
which is rendered clearly visible to the observer by the
whirling fragments of broken cloud floating below the

CAUSES OF THE FORMATION OF CLOUDS 67

main mass. The squalls that at times accompany
thunderstorms are caused by the descending cooled air
at B, and it is also this descending air which gives rise
to the mammato-cumulus referred to on p. 39.

2. By the Turbulence of Warm Air Flowing over Cold
Surfaces.—lf warm air flowing from a region of high
temperature reaches a region where the temperature of
the land or sea is lower than its own, the air layers in

contact with the surface will become cooled, and the
eddy-motion, which is caused by the friction met with
by the wind in flowing over the earth’s surface, will ex-
tend this cooling upward by mixing up the air, thus
producing an inversion of the normal lapse-rate from the
surface up to a height which depends upon the time and
distance the current has been moving. Eddy-motion is
most active during bright windy weather, and its action
can easily be seen by observing how rapidly the smoke

68 CLOUDS

from factory chimneys is tossed about and dispersed in
the air by the atmospheric eddies. The shimmering
seen along the horizons on land and sea is another
illustration of their activity. Major G. I. Taylor,
F.R.S., who has investigated eddy-metion, has explained
the frequent and dense fogs met with off the Banks of
Newfoundland as being due to the above cause.’ He
found that the warm air-current coming off the land to
westward and also that coming from over the warm Gulf
Stream water became cooled in their lower layers by
contact with the water of the cold Arctic current from
Baffin’s Bay, and that the cooling was propagated up-
wards by the eddy-motion, whereupon the moisture
contained in the warm air became condensed and formed
a dense fog, which after all is but a cloud resting on the
surface of the earth. An interesting development of
this method of cloud-formation is often met with at
Aberdeen during foggy weatherinsummer. An easterly
wind leaving the shores of the Continent as a warm
wind sometimes has its lower layers cooled by the water
of the North Sea-sufficiently to cause fog on our shores ;
or else the fog, which has been formed by the warm air
lying for some time nearly stagnant over the North Sea,
may drift westward under the influence of an easterly
breeze. There is formed in such cases an inversion
reaching upwards from the surface for a few hundred
feet, the air gradually becoming warmer upwards for
this distance before the normal falling-off of temperature
with height occurs. When, however, this fog-laden air
crosses the shore-line, the much warmer land at once
commences to heat the surface layers again, and the

'G. I. Taylor, M.A., Report of S.S. “ Scotia” to Board of Trade,
IQI4.

CAUSES OF THE FORMATION OF CLOUDS 69

inversion begins to change back into the normal lapse,
with the result that the fog evaporates near the surface
and passes overhead as a low stratus cloud. Farther
inland the inversion disappears altogether and the cloud
dissipates also, Sir Napier Shaw' suggests eddy-motion
as the cause of certain clouds in easterly and northerly
winds of long “fetch,” and some forms of cloud which
are very common on the borders of anticyclones in
easterly and south-easterly weather at Aberdeen can
scarcely owe their existence to any other cause, for
easterly currents in such types of weather maintain their
easterly direction to comparatively great heights, while
the clouds referred to are very low. There can therefore
be no warm moist current from another direction to assist
in their formation.

Clouds formed by this method are generally of strati-
form or nimbus types, and are described in Chapter VI.

3. Local or General Fall of Pressure may, by causing
expansion of the air, produce cloud if the air is near the
point of saturation. It is possible that the cirrus and
cirro-cumulus types may owe their formation, at least
partly, to this cause. The great rapidity with which
whole areas of sky may become covered with these
clouds, and the gradual thinning away at the edges of
the sheets, may be due to the flowing away of some
superincumbent air; and condensation produced by the
rarefaction would then take place in the damp areas.
Cirrus clouds are formed just below the base of the
stratosphere, where air-currents are probably flowing
smoothly and horizontally, and this might account for
their stream-like banded arrangement.

1 Sir Napier Shaw, F.R.S., ‘‘ Manual of Meteorology,” Part IV,
chap. v., 1919.

70 CLOUDS

4. The Mixing of Two Masses of Saturated Arr of
Different Temperatures.—Clouds which may be pro-
duced by this method are likely to be only thin stratiform
layers or films resembling haze. A cubic foot of air at
a temperature of 64° F. contains, when saturated, 6°6
grains of water-vapour; the same volume at 50° F.
contains 4°1 grains. If both volumes are mixed the
temperature of the mixture becomes 57° F’. and the water-
vapour present will then be 5°35 grains per cubic foot.
But air at 57° F. can hold only 5°2 grains per cubic foot,
therefore the excess of 0°15 grain must condense as a
cloud. But condensation will warm the air somewhat,
so that only part of the excess will become cloud.

5. Extenswe Cloud-layers and the Effect of Radwation.
—Clouds that appear in widely extended sheets, some-
times covering vast areas, seem to be closely connected
with the stratified conditions found in the atmosphere.
Balloons with recording instruments attached to them
have made many ascents into the free air, and their re-
cords have demonstrated that the atmosphere is really
built up of a series of layers of air of widely differing
temperatures and humidities, and also of varying veloci-
ties and directions of motion. Our knowledge of these
conditions has been greatly added to by Captain C. K.
M. Douglas, R.A.F., who, during the last year of the
war, was attached to the Meteorological Section of the
R.E. in France. Captain Douglas, in his aeroplane,
which was fitted with thermometers and a camera, ob-
tained a series of observations of the meteorological con-
ditions in the neighbourhood of clouds, in addition to a
large number of beautiful photographs of the clouds as
seen from above. A few of these wonderful cloudscapes
are shown in Plates 38 to4o. One of the most important

CAUSES OF THE FORMATION OF CLOUDS 71

results was the confirmation of the intimate connection
between strato-cumulus clouds and inversions of tem-
perature, a fact which had previously been established
by kite and balloon observations.’ On passing through
a layer of strato-cumulus Captain Douglas found that
the temperature, which had been falling as he rose to the
cloud level, often began to rise just above the clouds and
the increase extended upwards for several hundred feet.
Increases of temperature of from 1° F. to 10° F. were
met with, and in some exceptional cases the values were
even larger, an increase of as much as 15° F. having
been recorded in 1000 feet of ascent. The cause of
these inversions is still a somewhat debatable point ;
eddy-conductivity may, if vigorous, by mixing up the
air, produce heating at the surface and cooling at the top
of the layer in which it is operative, and at the same time
will carry up moisture which will condense at the level
where the air becomes saturated. But the fact that
Captain Douglas frequently found the warmer layer
above the clouds to be also much drier than the layers
below seems to suggest that the inversion may some-
times be due to the slow descent of air, which is warmed
in the process of descent. It is important to note that
though ascending air may have its rate of cooling by ex-
pansion reduced to about 3° F. in 1000 feet owing to
the access of heat liberated by condensation and rain-
formation, descending air, on the other hand, must be
warmed by compression at the adiabatic rate of 53° F.
in 1000 feet and such air will therefore always be

1W. H. Dines, F.R.S., ‘‘ The Vertical Temperature Distribution
in the Atmosphere over England,” “ Phil. Trans. Roy. Soc.,” Series
ever 2ti, Dp. 25%.

72 CLOUDS

relatively dry.'. The thickness of the layers of strato-
cumulus is sometimes remarkable. Captain Douglas on
frequent occasions has found them to be over 4000 feet
thick.”

During quiet anticyclonic weather in winter-time,
sheets of cloud of the stratus type, of from 500 to 2000
feet in thickness, often form at low levels. Mr. W. H.
Dines is of the opinion that these cloud-sheets are most
probably caused by the loss of heat due to radiation from
the air layer in which they form.’ |

The elevation of a saturated layer or the reduction
of its pressure will, by cooling, produce a cloud-sheet,
and it is very probable that the widespread alto-stratus
in cyclonic areas is formed in this manner. Captain
Douglas has found this cloud generally to be composed
of snow.

Another frequent cause of sheets of high cloud is the
undercutting of warm damp currents in the upper atmos-
phere by colder ones from other directions, much the
same process in fact as that which gives rise to the sur-
face “‘line-squall”.. The warm air is forced upwards by
the cooler current, the result being a wedge of warm
moist air spreading above the cold air, and thinning out
at its boundary into one of those clearly-defined sharp
edges, sometimes hundreds of miles in length, that are
so noteworthy a feature of the upper and intermediate

1W. H. Dines, F.R.S., loc. cit., ‘Phil. Trans. Roy. Soc.,” Series
A> Vol 2hits p263:

*For this and many more interesting details see the papers by
Captain Douglas in the “Journal of the Scottish Meteorological
Society,” 1916-17-18, and also “‘Symon’s Meteorological Magazine,”
December, 1919, and January, 1920.

? See “ Manual of Meteorology,” Part IV, p- 49, footnote.

CAUSES OF THE FORMATION OF CLOUDS 73

cloud-layers (see Plates 108, 14.8). These sharp edges
very frequently accompany the passage of the trough
line of a depression, and are often revealed by the
clearing away of the lower cloud at the moment of the
passage of the trough line.

The beautifully waved structure seen in nearly all of
the layer-types of clouds from cirrus downward to strato-
cumulus is caused by the propagation upwards or down-
wards of the wave-motion that is produced by the flow-
ing of air-currents of different velocities and directions
over each other, and was first explained by Helmholtz.
Exactly similar patterns are formed by water-currents
rippling over sand, by the wind rippling the surfaces of
the ceolian sand-dunes, and also by the wind on the sur-
face of drifted snow. It is not always that the wave-
cloud occurs at the shearing plane between two currents,
but as the wave-motion is propagated vertically it follows
that wherever there may exist a layer near the point of
saturation, any lifting of that layer will produce cooling
and condensation. Consequently, where the crests of
the waves occur, such a layer will be lifted and waves
of cloud will be formed, while in the wave-hollows the
layer will be depressed and the compression will prevent
condensation, thus producing the intervals of blue sky
between the cloud-bands. In addition to these easily
visible waves there sometimes occur systems of very
small ripples which can be seen only by telescopic aid.
It is a well-known fact that if a flame—such as a candle
flame—be held in the path of a ray of sunlight, the rising
of the heated air can be rendered visible by letting the
shadow of the candle fall upon a piece of white paper,
or even by holding the candle between the eye and the
sun. At Aberdeen a telescope with a dark-red glass

74 CLOUDS

attached to the eyepiece has often been directed at the
sun when bands of thin high cloud have been passing
over the sun’s disc, and on some occasions trains of tiny
ripples lasting for only a second or two have been seen
to form and disappear in rapid succession within the
cloud-band. In these cases it has been the moving air
itself that formed the ripples, and the air-current pre-
sented an appearance exactly like that of a swiftly flowing
rippled current of water.

The lenticular cloud-banks are also due to a wave-
motion. It has been mentioned that they appear in air-
currents flowing across mountainous country, and there-
fore the obstruction offered by the mountain masses to
the flow of the air will tend to elevate somewhat the air-
layers above, when, if the latter are damp, condensation
will occur at the points where the elevation takes place.
Such a wave will be stationary or may move forward
very slowly, and may set up similar waves in front of it ;
and the swiftly moving air will pass across this stationary
wave-system, condensation occurring as it crosses the
crest, and evaporation as it descends into the trough.
The peculiarities of these clouds as described in Chapter
III will thus be accounted for.

CHAE ER CN,

CLOUD DISTRIBUTION, HEIGHTS, DIRECTIONS, VELO-
CITIES, AND FREQUENCIES,

Cloud Distribution.—The amount of cloud present in the
sky varies very greatly with locality. Angot' states that
in the same latitudes the amount of cloud is usually
greater over the oceans than in the interior parts of the
continents, while over desert areas the amount of cloud
is negligible. Likewise there is more cloud on coasts
which are subject to ocean winds, and less on those
where the prevailing wind blows off-shore. Thus it is
that Scotland, Ireland, and the north-westerly and
central parts of England are more cloudy than is the
south-eastern portion of England, the amount of cloud
in our islands ranging from about 70 per cent. in the
former districts to about 60 per cent. in the last named.
The cloudiest part of the earth in the northern hemi-
sphere is found in a belt stretching from the coast of
Labrador round the south of Greenland, thence pass-
ing between Iceland and Scotland round the north of
Norway below Spitzbergen. A similar belt crosses the
north Pacific below Kamchatka and Alaska.

In our islands there is a greater amount of cloud
during the winter half of the year, and the daily range

1A, Angot, “Traité élémentaire de météorologie,” Book III,
chap. ii., paragraph 68.
(75)

76 CLOUDS

shows a maximum of cloud somewhere in the middle of
the day, and a minimum very late at night. But in
other parts of the world the annual variation in cloud
amount differs greatly.. For example, at high mountain
stations the maximum of cloud occurs in summer and
the minimum in winter, because the clouds being rela-
tively lower in winter are then found chiefly dedow the
level of the station. Then again, stations which experi-
ence a wet summer season, as do those on the west
coast of India, have a decided maximum of cloud in
summer. |

Cloud Feights.—In Chapter II there were given the
average heights of each cloud type. But, as might be
expected, the average height of any particular type of
cloud depends upon the latitude; a greater height is
found in the equatorial regions and a lesser one in polar
latitudes. This fact was well brought out by the inter-
national cloud observations made throughout the world
in the years 1896-97. There was also proved to exist
an annual variation in heights : in summer the clouds are
higher than in winter, the actual curves showing this
annual height-variation have been given by Bigelow.'
Asa rule the maximum height occurs in summer, and
the minimum in winter or early spring.

Directions of Cloud-motion.—In our latitudes in the
northern hemisphere the prevailing surface winds are
from the south-west, but higher up in the atmosphere
the winds become steadily more westerly. That this is
a fact has been demonstrated from the results of pilot
balloon ascents made at the Meteorological Office obser-

* Bigelow, “ Report on International Cloud Observations,” 1900.
“United States Weather Bureau,” quoted in McAdie’s ‘Principles —
of Aérography,” 1917.

“CLOUD DISTRIBUTION, ETC. 77

vatories and also at Petersfield by Captain Cave.’ It is
also borne out by the directions from which the clouds
move. An analysis of the directions of cloud-motion
from observations made at Aberdeen with the Fineman
nephoscope is given in the supplement to the “ Geophysi-
cal Journal” * for 1916. The observations cover a period
of five years from 1912 to 1916, and the results are
shown in the vector diagrams in Fig. 14, where eight
types of cloud are represented. The two other types,
nimbus and stratus, are not amenable to measurement.
But in the types analysed it is perfectly obvious that by
far the greatest proportion of the clouds come from some
westerly point, while the paucity of easterly directions
is remarkable. Strato-cumulus shows the least bias, and
is somewhat resembled in this respect by alto-cumulus,
and rather less so by cirro-cumulus. Cirro-stratus and
alto-stratus do not convey an altogether correct impres-
sion, because both these types occur very frequently in
front of a depression and should therefore show a con-
siderable grouping between south and south-west in ad-
dition to the westerly maximum. But when in front of
a low-pressure area, both these types are very uniform
and do not give sufficient detail to enable measurements
to be made of their motion, except on rare occasions ;
the diagrams given are the results of measurements
chiefly of the types which show a reticulated structure,
and which are usually found in westerly weather.

Most of the cloud-types exhibit some inherent peculi-

1C, J. P. Cave, M.A., “The Structure of the Atmosphere in
Clear Weather,” 1912.

2« Geophysical Journal,” 1916. Meteorological Office, 227 (d),
“ Annual Supplement,” p. 80. The observations and vector dia-
grams are herein given in much greater detail.

78 CLOUDS

arity ; thus cumulus and cumulo-nimbus are found chiefly
in the north-westerly quadrant, the cumulo-nimbus hav-

Cirro-StratuS

OF ee ee eee eee

Cirro-Cumulus. Alto-Cumulus.

Alto-Stratus. Strato-Cumulus.

Cumulo-Nimbus.

Fic. 14.—Percentage frequency of directions of cloud-motion at Aberdeen. The
outer circle indicates 25 °/, of total number of observations; the inner one
indicates to °/,. The straight lines show the percentage of observations of
cloud from each direction. j

CLOUD DISTRIBUTION, ETC. 79

ing a more northerly bias than the cumulus. Had it
been possible to measure nimbus and stratus, the former
would have shown a strong maximum between south-
east and south-west, while the latter would have been
distributed mainly between south-east and north-east,

Cirrus shows a decided maximum from the west-
ward, but it is noteworthy that all the cirrus types are on
a few occasions found to move from some easterly point,
so that easterly winds may at times reach up almost to
the base of the stratosphere.

The sequel to these distributions is given in Chapter
Ve

Using the international cloud observations as a work-
ing basis, Hildebrandsson and Teisserenc de Bort’ have
formulated a most suggestive and interesting explana-
tion of the general atmospheric circulation on the earth.
In equatorial regions the cirrus clouds showed an easterly
current of high velocity in the upper air.

Velocities of the Clouds.—The velocities of the clouds
were also measured at Aberdeen simultaneously with
their directions, and the results obtained demonstrated
that in general the apparent velocities were greater in
winter than in summer, very markedly so in the case of
the “convection” type of clouds, cumulus and cumulo-
nimbus. This must be due partly to the stronger winds
in winter, but partly also to the lower altitude of the
clouds at that time of year. The cloud-velocities mea-
sured were their azgular velocities, and these would have
to be modified according to the actual heights of the
clouds in order to obtain their actual velocities. If the
average angular velocities as found by observation are

1 Hildebrandsson and Teisserenc de Bort: “Les bases de la
météorologie dynamique,” chapter iv, 1903.

80 * CLOUDS

treated in this manner, using for the average cloud-
heights those that are given in the International Classi-
fication, then the following average actual velocities
in miles per hour are obtained.: cirrus 60, cirro-stratus
60, cirro-cumulus 45, alto-cumulus 30, alto-stratus 35,
strato-cumulus 22, cumulo-nimbus (central mass of cloud)
35, and cumulus 31. Some of the individual velocities
obtained for cirrus in a westerly current were very high,
for even assuming the height of the cloud to have been
only 23,000 feet instead of the 30,000 feet given as the
average, the velocities not infrequently reached 100 or
130 miles per hour.

Frequencies of the Cloud-types.—Though any par-
ticular type of cloud may occur at any time, yet there
exists some slight periodicity in the appearance of certain
types, notably in the cases of the cumulus and cumulo-
nimbus, which show a seasonal range as markedly as
they do a diurnal one. The analysis that has just been
quoted, together with another one which deals exclusively
with the frequency of cloud-type,’ both tend to show
that at Aberdeen these two cloud-types are more fre-
quent at the equinoxes than in midsummer and have
an absolute maximum in April. On the other hand, at
Epsom in the South of England, an analysis made by
S. C. Russell* shows a definite maximum for these
types in midsummer. Both the Aberdeen and Epsom
analyses agree in showing strato-cumulus to be more
frequent in winter than in summer, but Epsom has the

1 “An Analysis of Cloud Distribution at Aberdeen, 1916-1918,”
by G. A. Clarke. M.O. Professional Notes, No. 9, 1920.

» Spencer C. Russell, “Results of Monthly and Hourly Cloud-
form Frequencies at Epsom, 1903-1910,” “Q.J.R. Met. Soc.,”
October, 1913.

CLOUD DISTRIBUTION, ETC. ot

maximum for the cirrus varieties during the summer
while Aberdeen records a late autumn to winter maxi-
mum for them. The existence of such differences be-
tween two stations almost at the two extremities of
Great Britain demonstrates that the cloud conditions
vary very greatly with locality.

This fact is likely to have a considerable bearing
upon aerial navigation, for, in the event of transport by

Fia, 15.—Distribution of cloud at meteorological stations.

air assuming any considerable dimensions, it will become
necessary for the cloud conditions at all stations to be
known. Recently the Meteorological Office in its
“Upper Air Supplement to the Daily Weather Report”
~ has been giving maps indicating the cloud at a number
of stations. Fig. 15 showsa modification of these maps,
two examples being given showing different conditions.
In (a) there is shown a case where a good deal of high
cloud (7000 feet and Berra) is present, and only a

82 CLOUDS

small quantity of low cloud (below 7000 feet) though in
the south-east of England it is of unfavourable type,
that is to say, the cloud is either of the convectional
order (cumulus or cumulo-nimbus), which makes the air.
very “bumpy,” or there is fog present or else rain is
falling and destroying visibility. In (4) there is much
low cloud, but little convectional cloud or rain is present.
In the maps the proportion of the circle covered with
each shading shows the proportion of that particular
cloud-type visible at each station. The solid black
shading indicates convectional cloud, rain, or fog, the
hatched shading shows low cloud (below 7000 feet), and
the unshaded portions indicate either clear sky or high
cloud (above 7000 feet).

CHAPLE RIV 1,
THE ASSOCIATION OF CLOUD WITH WEATHER TYPE.

From the earliest times it has been customary for ob-
servers of the weather to associate certain cloud forms
with certain kinds of weather, and by patient observa-
tion and correlation much knowledge has thus been ob-
tained which is of value. It has been established that
certain clouds or combinations of cloud-types are liable
to recur in similar weather conditions; of these the
general sequence of cloud in acyclonic depression is the
most familiar. But even the same clouds do not always
mean the same weather in all localities. Angot, in his
treatise on meteorology’ mentions that cirro-cumulus
in our islands is usually a good sign, while in Italy it
portends the reverse; and that cirrus, when light and
motionless, accompanies very fine weather, but if seen
in swiftly moving bands is generally the forerunner of a
storm. If, however, weather-type, as determined by
the pressure distribution, is taken into consideration in-
stead of the actual weather, it is found that certain cloud-
forms are more prevalent than others according to the
weather-type in question. In the diagrams which ac-
company this chapter there are shown the British Isles
and some neighbouring parts of the continent. Over
each map there is drawn a certain arrangement of lines
indicating where barometric pressure is the same ; these

1 Loc. cit., Book B, chapter i, par. 111.

(83)

84 CLOUDS

are the “‘isobars ” that have been referred to in Chapter
IV, and according to the direction in which they lie,
and to the relative positions of the high and low pressures,
the prevailing winds can be deduced from them. It is
convenient to refer to the wind as a method of indicating
the weather-type. Thus an arrangement of isobars
running north and south with high pressure to the west
and low to the east, will have in general a northerly
wind blowing across our islands, and can be termed
‘northerly’ weather. In the maps the arrow (or
arrows) indicate the direction of the wind, and the num-
bers at the top of the isobars are conventional readings
of the barometer.

The following details of cloud likely to be found in
each weather-type are from observations at Aberdeen
only and may therefore to some extent be influenced by
locality.

South-easterly Lype.—High pressure lying gener-
ally over Scandinavia and low pressure to the south-
west of Ireland. Wind varying between S.E. and E.
Cloud is almost always of the stratus type, sometimes
complete, when it may degenerate into a nimbus pall
with light drizzle, at other times broken into cumuliform
masses. It is often. very persistent. When moving
from about east, these masses very strongly resemble
sheets of strato-cumulus, and are apparently the clouds
referred to by Sir Napier Shaw’ as due to eddy conduc-
tivity. They are extremely difficult to classify satis-
factorily ; possibly a name like “ eddy-cumulus” might
best describe them. The accompanying weather is

rather variable but generally cloudy to dull with mist and
drizzling rain.

‘See Chapter IV, page 69.

ASSOCIATION OF CLOUD WITH WEATHER 85

C. Northerly Type. D Southerly Type.

Fic. 16,—Examples of weather-types. Isobars marked in inches and millibars,

86 CLOUDS
North-easterly Type.-—High pressure to N.W. of

Scotland, low over eastern France. Wind varying
between E. and N.N.E. The cloud, when the wind is
more easterly, is of cumuliform stratus type similar to that
mentioned above, but as the wind backs more to the
northward this gradually gives place to rather larger
masses, exactly like the ordinary cumulus in appearance,
or sometimes resembling small cumulo-nimbus, when
very slight showers may fall from them. Above them
is a layer, more or less complete, of normal strato-
cumulus, into which the summits of the lower cumulus
and cumulo-nimbus at times reach. The accompanying
weather is mostly fair, and visibility as a rule is very
good.

Northerly Type.—A large anticyclone to the west of
Britain and either a single or multiple low to the east of
the North Sea. Wind from N.N.E. to N.W. Charac-
terised almost exclusively by small to moderate-sized
cumulo-nimbus clouds, of which the “anvil” portion
usually forms the major part; the clouds move from a
direction somewhat more easterly than that of the sur-
face wind, especially the upper “anvil” part, which
usually develops into a layer of small strato-cumulus or
alto-cumulus as the wind dies down. Weather is very
squally and cold; with very frequent showers of rain,
hail, sleet or snow.

Southerly Type.—The isobaric conditions are exactly
the reverse of the above. Wind is between S.S.E. and
S.S.W. This type shows a double cloud character. If
the British Isles come more under the influence of the
anticyclone the cloud is chiefly a low broken stratus,
sometimes with strato-cumulus above it, but under the
influence of the low pressure the dominant types are

ASSOCIATION OF CLOUD WITH WEATHER 87

cirro-stratus and alto-stratus with ‘‘scud”’ stratus below.
Nimbus may follow if the low pressure advances farther
over our islands but the sequence of cloud in this case
is dealt with under cyclonic weather. The usual weather
in this southerly type is dull, gloomy, misty and raw,
generally with some rain.

South-westerly Type.—The isobars often stretch from
the Azores to Norway straight across our islands, or
else are curved as indicated in the figure. Pressure
is high over the continent, low out on the Atlantic, and
the wind blows from some point between S.S.W. and
W.S.W. The clouds comprise all the varieties of upper
and intermediate clouds from cirrus and cirro-cumulus
to alto-cumulus and strato-cumulus. Occasionally some
slight cumulus and broken stratus may also be present.
At Aberdeen all the intermediate and higher clouds
above-mentioned appear grouped in lenticular masses,
which is a local characteristic, as was described in the
previous Chapters III and IV, and the weather found
at Aberdeen under these conditions is very warm and
squally, probably also a local modification of the general
fine dry weather of the anticyclone. Farther to the west
the weather may come under the influence of the
Atlantic low pressure and there may then be more low
stratus and nimbus cloud.

Westerly Type.—High pressure lies to the south, over
the Bay of Biscay and Spain, and the isobars run from
west to east across our islands. Low pressure is found
to the north between Iceland and Norway. Wind at
the surface blows from between W.S.W. and W.N.W.
The chief characteristic clouds of this weather-type are
finely formed bands of cirrus and high cirro-cumulus,
changing occasionally into sheets of cirro-stratus and

88 CLOUDS

1019 mb.

10pm:

E. South-Westerly Type. F. Westerly Type.

G. eres cerly Type Variation. H. Cyclonic Type.

Fic. 17.—Examples of weather-types. Isobars marked in inches and millibars,

ASSOCIATION OF CLOUD WITH WEATHER 89

alto-stratus. Sometimes they are accompanied by lenti-
cular clouds of the south-westerly type, and there may
also be some strato-cumulus present, but only very slight
quantities of cumulus are seen. The weather is usually
fair.

Whe next stype (Misy17 (c)) is a variant of ‘the
westerly type, where the high pressure is situated out
on the Atlantic to the south-west of our islands, and
spreads across them in a projecting shoulder. A succes-
sion of depressions is passing in an E.S.E. direction
across the northern part of Great Britain towards the
continent, and the shoulder of the high thus becomes a
region of frequent surges of low pressure, It is a very
common distribution, and has been mentioned here
because it seems to be a region where the cloud is
chiefly of the strato-cumulus type. A fair amount of
cumulus is also present, and in this respect the region
resembles the rear part of a cyclone.

Cyclonic Type.—Owing to its comparatively rapid rate
of travel, this type shows a sequence of cloud-forms that
has become very familiar to most people. Inthe diagram
the large dotted arrow shows the path and direction of
the movement of the depression, and the small solid
arrows the various winds found in the different sectors
of the cyclone. The dotted double line represents the
“trough” of the depression—the line along which the
pressure commences to rise again. Considerably in
front of the cyclone, sometimes even before pressure
has begun to fall, there will be seen banded cirrus,
moving from the S.S.W.; this will gradually close up
into cirro-stratus, which in turn will thicken into alto-
stratus as condensation increases. Then as the mercury
falls at an increasing rate the nimbus cloud with rain

fete) CLOUDS

follows. In the right front sector (A) the nimbus its
usually dense and ragged with “flying scud” below it ;
in the left front sector (B) it is more uniform in appear-
ance and the rain is usually less heavy but very persis-
tent. Occasionally the rain continues to fall for a short
time after the pressure has begun to rise, but at other
times the passage of the trough line over a station is
accompanied by the cessation of the rain and the com-
mencement of the clearing of the nimbus cloud. When
this takes place cirro-stratus or alto-stratus may still be
seen above the broken lower cloud, but this high cloud
also disappears before long, and in the rear of the dis-
turbance the cloud-forms are usually strato-cumulus,
cumulus, and fracto-cumulus in the right rear sector (C)
and strato-cumulus with fracto-cumulus and small cumulo-
nimbus in the left rear (D).

In the cases where the cyclone is elongated into what
is known asa ‘“V” shaped depression the passage of
the trough of the “ V ” is often accompanied by conditions
similar to those of a line-squall, and the cloud in this
case is very thick and heavily mammillated, and of the
cumulo-nimbus character. The “clearing squall” often
shows an upper cloud-sheet with a straight edge, still
“moving from the south-westward, though the surface
wind has become north-westerly.

The nimbus cloud in small secondary depressions is
often mixed with the cumulo-nimbus type and accom-
panied by thunderstorms.

The foregoing are, of course, simply generalisations
upon the distributions of cloud-types. Weather-type is
perpetually in a state of flux, and therefore the cloud-
character is one of constant mutation. Therein lies its

ASSOCIATION OF CLOUD WITH WEATHER gi

greatest interest, and of the problems it presents to us
many of the most important still remain unsolved.
Those who would become acquainted with these prob-
lems are recommended to turn to the pages of Sir
Napier Shaw’s book “ Forecasting Weather”.

SECTION II

NOTES ON PLATES OF PHOTOGRAPHS OF CLOUDS.

In the following plates the clouds have been so arranged that the
highest varieties come first, and all the other types follow in the
order of their heights.

Opposite each illustration there is given in italics the correct
classification for each cloud according to the international nomen-
clature ; and the classification is followed by a detailed description
giving the salient features of each cloud picture. It is hoped that
by these means observers may find a ready guide to the classification
of such clouds as they may from time to time observe.

For the benefit of those to whom cloud photographs may not
be quite familiar, and who may therefore experience some slight
difficulty in correctly translating them, it may be mentioned that the
blue of the actual sky appears as varying shades of grey to black in
the pictures according to the methods employed in taking the photo-
graphs. In all the pictures of cirrus, cirro-stratus, cirro-cumulus,
and false cirrus, as well as in some thin varieties of alto-cumulus, the
cloud appears as white on a black sky. In the case of the heavier
lower clouds, and of the lenticular cloud-banks, the clouds show
definite shadows, which appear sometimes darker than the sky tone,
and sometimes lighter, but all the examples shown have been chosen
with a view to eliminate any possible confusion of cloud-shadow
with sky. Alto-stratus and stratus being uniform cloud-sheets have
no blue sky visible, and therefore appear in tones of grey as they do
in nature.

(95)

96 CLOUDS

PLATE 5.
5a. Cirrus. Tufts of cirrus with extremely delicate thread-like
continuations. The cirrus is moving in the same direction as that

in which the lines are lying; the tufts preceding, the tails following.

58. Cirrus, Photograph taken a few minutes later of the same
bands of cirrus as shown in the above view. The tails have become
much denser and longer, but, though perspective has altered the
relative positions of the tufts, yet there is no difficulty in recognising
the same tufts in both photographs.

PLATE 6

CLOUDS 97

PLATE 6.

6a. Cirrus. An extremely fine example of long threads of cirrus,
ending in some cases in dense tufts. This formed part of a long
band of cloud, and the threads were lying ¢ransversely across the
band.

6p. Cirrus, Extensive curved threads, forming part of a huge
plume of cirrus. The-ends of the threads show the familiar tufts,
though not so markedly as in the foregoing case,

98 CLOUDS

PLATE 7.

za. Cirrus. Very dense masses of cirrus which give evidence of
considerable vertical thickness. A type which showed very rapid
changes of form and eventually became cirro-cumulus. Thread
structure is very slight and ill-defined.

7B. Cirrus. Isolated, rather heavy threads of cirrus which show
at their lower extremities a tendency to break up into faint flakes.
This process continued until the whole band became ordinary cirro-
cumulus,

PLATE 7

ee ee ey

PLATE 8

CLOUDS 99

PLATE 8.

8a. Cirrus. Long and very sharply defined threads along which
there occur nuclei from which other threads spread outwards more
or less transversely to the main threads. This form often develops
into the wave-like arrangement shown in Plate 128.

1

8p. Cirrus. Detail of a portion of a long cirrus band. A rather
unusual variety of the tufted forms. Throughout the length of the
band there was considerable variation ; in some portions the filaments
were curved but delicate in structure, and in others the cirro-macula
or “speckle-cloud ” form was developed.

100 GLOUDS

PLATE 9.

ga. Cirrus. A very interesting case where cirrus cloud is pres-
ent at two levels. The higher of the two is represented by the
small patches lying nearly horizontally at the right-hand side of the
picture. The lower cirrus is represented by the converging bands
running from top to bottom of the picture. These latter bands
show two different characters, the right-hand one being inclined to
the ‘‘speckle-cloud” type while that on the extreme left consists of
very delicate transverse threads. Both cloud layers were moving
away from the observer towards the direction shown by the
convergence of the bands. The lower cirrus cloud was moving
apparently about half as fast again as the upper layer.

gs. Cirrus. The same cloud as the above, but photographed
some ten minutes later. The upper cirrus layer is here shown in
greater extent, while the lower layer is reduced to a single band
which exhibits well-marked wave-motion across its length. The
change in the angle at which the band is lying is, of course, the effect
of perspective ; the band lay to the right of those shown in a above,
and converged to the same vanishing-point.

PLATE 9

a +s eas

PLATE 10

—
W

CLOUDS IOI

PLATE tro.

10A. Cirro-stratus. ‘Type formed by the coalescence of a multi-
tude of fine cirrus threads. Sheets of this type are found spreading
out in advance of a depression. Note the increase in the density of
the cloud at the left of the picture.

top. Cirro-stratus. Part of a widely spread sheet of very dense
cirro-stratus some twenty thousand feet or more above the small
cumulus clouds along the horizon. ‘The straight edge, which lay to
the westward, was very remarkable, inasmuch as it persisted for at
least twelve hours, the cloud moving in the direction indicated by
the line of the edge, that is, from left to right, so that if the cloud
had been moving even at the comparatively moderate velocity of 40
miles an hour, the edge must have been at least 500 miles in length.
As time went on the cloud grew denser, finally becoming alto-stratus,
and early next day a fierce blizzard broke over Aberdeen and the
north of Scotland generally.

102 CLOUDS ;

PLATE 11.

11a. Cirro-stratus. Actually a very fine example of cirro-nebula,
the cloud being so delicate as to be practically invisible. ‘The sun is
hidden by the dark mass of strato-cumulus at the bottom of the
picture, and the ice-crystals, of which the cirro-nebula is composed,
are producing in the cloud the bright arch of light which is part of a
very fine solar halo. The angle of view of the camera-lens was in-
sufficient to enable the whole halo to be included in the photograph.

11B. Cirro-stratus. Broad bands of cirro-stratus edged with fila-
ments of cirrus, and having above them a narrower band showing
the rippled structure characteristic of cirro-cumulus. A rather dense
variety of cirro-stratus, showing no thread-structure except at the
upper edge.

PLATE 12

CLOUDS 103

PLATE 12.

12A. Cirrus. The form of cloud that shows an intermediate con-
dition between the cirrus and cirro-cumulus types. Threads and
filaments are mixed up with small globular masses that strongly re-
semble little balls of cotton-wool and are intensely white, This
form is often found arranged in long bands during westerly weather.

128. Cirrus. The waved form that usually develops from cirrus
of the type shown in Plate 8a. It is very evanescent, and has
been called “change cirrus” by Clayden because it is in a state of
continual flux between the cirrus thread form and the cirro-cumulus
type. When extensive, it is probably the most beautiful variety of
cirrus.

104 CLOUDS

PLATE 13.

13a. Cirro-cumulus. An example of the variety termed “ speckle-
cloud”; the globules of cloud are excessively minute toward the
edges of the sheet, where they become mixed with very delicate
threads of cirrus.

13B. Cirro-cumulus, Part of a long band of small cirro-cumulus,
fringed with a line of cirrus plumes. Both types belonged to the
same system, the gradation from the one to the other being visible
in various places along the edge of the cirro-cumulus. In a short
time the cirrus became transformed into cirro-cumulus.

PLATE 33

See

¢

. n ~
. * + ‘

4 «

7

CLOUDS 105

PLATE a4.

144. Cirro-cumulus. Intensely white high cloud in radiating
bands. At the upper left-hand corner it will be seen that the cloud
is Opening up into the familiar globular masses of cirro-cumulus.
The height of these bands was probably about 20,000 feet, while the
broken cumulus below was not more than 5000 feet high.

148. Cirro-cumulus. The same bands mentioned above as they
appeared some time later and in a different part of the sky, which
accounts for their different apparent orientation. The character of
cirro-cumulus is now well-developed, and wave-motion can be seen
in the bands both longitudinally and transversely. ‘These bands
were not less than 250 miles in length.

106 _ CLOUDS

PLATE 15.

15A. Cirro-cumulus. Part of a sheet of fine globular cirro-
cumulus which exhibits double undulation, the transverse system
being rather irregular. This example formed very rapidly from a
previously uniform cloud-sheet.

158. Cirro-cumulus. The waved type of cirro-cumulus, the
cloudlets being formed by a single wave-system. Such clearly
defined waves are usually very transient, but this example was of
fairly considerable duration, the cloudlets gradually evaporating at
the upper edge of the sheet. Some very dark strato-cumulus
occupies the lower part of the picture.

B

CLOUDS 107

PLATE 16.

16a. Cirro-cumulus. A very unusual variety wherein the cloud-
lets are massed into small irregular rings with an open centre, thus
giving the grouped arrangement a “rosette” appearance. This
arrangement may be seen occasionally in most globular forms of
cirro-cumulus, but seldom so well-marked as in this case. The
cirro-cumulus is also of a heavier type than usual.

168, Cirro-cumulus. Very sharply.defined parallel bands of
cirro-cumulus, showing regular cross-striation in the two central
bands, while the widest band has a flaked structure. In this case
the bands were of no great length, and lasted for only about an
hour; they were also quite local, the remainder of the sky being free
from cloud of that type, and showing only a few detached cumulus
clouds, some of which may be seen in the lower half of the picture.

108 CLOUDS

PLATE 17.

174. Alto-cumulus, Long waves of alto-cumulus which do not
exhibit any shadows on account of the thinness of the cloud. The
wave-system was well marked only in this portion of the cloud-layer,
the rest being more diffuse and irregular, as may be seen near the
top of the photograph.

178. Alto-cumulus. Another example of cloudlets without
shadows ; in this case the waves are sinuous instead of straight.
The thinness of the cloud-layer is very obvious, especially on the
left-hand side of the picture, but this was the case only at the edge
of the layer, at the point where the photograph was taken. Farther
to the right, the cloudlets were much denser and showed the normal
shadowed appearance.

‘PLATE 18

CLOUDS 0

PLATE 18.

18a. Alto-cumulus. A finely formed globular variety, the cloud-
lets of which are progressively smaller towards the lower edge of the
picture. This was actually the case, and is not entirely the usual
effect of perspective. The shadows are well shown, and the very
globular form is an indication of the cloud’s kinship to the “ turret-
cloud ” form shown in Plate 244,

188. Alto-cumulus. Arranged in flakes and small masses that
have a flattened appearance. This is a very common form of alto-
cumulus in the neighbourhood of Aberdeen, being much more
frequently seen than the waved types, and it is also frequently seen
at the advancing edge of a widespread sheet of heavier strato-
cumulus.

IIO CLOUDS

PLATE 10.

19A. Alto-cumulus. The most typical form of alto-cumulus,
all the characteristics of that cloud-type as described in the Inter-
national classification being present in this picture. The two wave-
systems crossing each other (‘‘ double undulation ”) are very, clearly
shown, and the solid character of the cloudlets is well brought out
by the slight shadows upon them,

19B. Adto-cumulus. The edge of a very compact sheet of alto-
cumulus wherein the separate cloudlets are compressed and fused
into each other. The lack of areas of blue sky between the cloud-
lets is compensated for by the light of the sun (which is directly
behind the cloud) throwing into strong light and shade the differing
thickness of the cloud-sheet, and thereby revealing the globular form
of the component cloudlets. At the right-hand edge of the cloud-
sheet there may be seen, against the blue sky, a large number of very
small incipient cloudlets ; these formed a fringe along the whole
length of the cloud-layer.

PLATE 19

PLATE 20

CLOUDS ILI

PLATE 20.

20a. Cirro-cumulus Lenticularis. Three lenticular-shaped banks
of cirro-cumulus clouds that have become so much fused as to lose
entirely their “cumulus” character. This is a very common occur-
rence with cloud of this type, and it is often only at the leeward edge
of the cloud-bank that the real character of the cloudlets becomes
evident. Only very slight traces of it can be seen in this picture,
but in the example below it is clearly shown.

20B. Cirro-cumulus Lenticularis. A portion of a cloud-bank of
this type, wherein there is shown a very beautifully rippled under-
surface, greatly resembling the surface of rippled water. The
cloudlets are passing outwards from the cloud-bank and moving
towards the right, where it will be seen that their cirro-cumulus form
is perfectly normal,

t12 CLOUDS

PLATE at.

21a. Cirro-cumulus Lenticularis. Here cirro-cumulus cloudlets
in long waves are massing together into a complicated system of
lenticular banks. The cloudlets were moving with a very high
velocity, but the cloud-banks remained practically stationary for
many hours, though their details and sizes were rapidly varying.
The great density of some of the banks is well shown in the lower
part of the picture.

21B. Cirro-cumulus Lenticularis. A very large compound len-
ticular cloud-bank with several small detached single clouds. This
picture is probably the most beautiful cloud photograph taken so far
by the writer, and shows a wonderful sunset effect upon the rippled
under-surface of the cloud-bank. The colours were very beautiful,
the sky itself varied from deep ultramarine at the top to emerald
green near the horizon, and the small dark masses of strato-cumulus
cloud were cold grey-purple with flame-coloured edges. The large
cloud-bank was of a deep ochre-yellow tint, while its edges and the
sunlit ripples on its under-surface were outlined in blazing gold.
The system of double undulation shown by the ripples is a specially
remarkable one.

PLATE ar

PLATE 22

CLOUDS 113

PLATE 22.

22A, Alto-cumulus Lenticularis, Lenticular cloud-banks at the
alto-cumulus level. Here again the individual cloudlet character is
suppressed, but may be seen in the bright portions of the two nearer
banks of cloud. In this instance, the banks overhead were com-
posed of clearly defined alto-cumulus.

228. Alto-cumulus Lenticularis. A series of these cloud-banks
well depicted at sunset. ‘The relatively great density of the central
parts is plainly to be seen, as also is the comparative thinness at
their edges. In the upper part of the photograph, the individual
cloudlets are seen passing out of one bank and entering another.
The proper method of distinguishing between cirro-cumulus and
alto-cumulus when they are massed into banks of this type is to
examine the individual cloudlets that are nearest to the observer,
for the banks themselves give very little indication of any difference
when they are at all dense or very distant.

II4 CLOUDS

PLATE 23.

23a. False Cirrus, Heavy masses of the so-called fa/se cirrus
surmounting some small cumulo-nimbus shower-clouds from which
rain may be seen falling. The false cirrus plumes will be seen to be
less clearly defined than those of the ordinary cirrus, though re-
sembling them in their wispy arrangement. They are also heavier
and denser in texture than the true cirrus plumes.

¢

23B. false Cirrus becoming Alto-cumulus. The cloud shown in
this picture had shown the ordinary cirrus character only a few
minutes before this photograph was taken. Here the last fibres of
the false cirrus will be seen on the extreme right in process of trans-
formation into the characteristic form of alto-cumulus. In the
central band, and in that at the lower left corner, the transformation

is progressively further developed, being complete in the last-
mentioned-band.

PLATE 23

CLOUDS 115

PLATE 24.

24A. Alto-cumulus Castellatus. ‘The ‘‘turret-cloud,” the form of
alto-cumulus that is so aften a precursor of thunderstorms. Its
resemblance to the ordinary cumulo-nimbus of lower levels is shown
clearly by its considerably developed vertical structure and by the
rounded protuberant upper surfaces of the cloudlets.

248. Adto-stratus. The; characteristic\ grey pall of intermediate
cloud through which the sun is.shining weakly. Some loose “scud ”
in dark ragged masses—the forerunners of the coming nimbus—may
be seen moving up below the alto-stratus. This is an example
wherein the alto-stratus is quite structureless.

116 CLOUDS

PLATE 25.

254. Strato-cumulus. The typical farm of this cloud. The
masses are heavily shaded, and tend to fuse into each other within’
the sheet, while at the edges of the sheet they appear separated, so
that the blue of the sky can be seen between them.

258. Strato-cumulus. A still heavier form than that -above
mentioned. This is the type of cloud-sheet that may cover the
whole sky for days on end, especially during the winter months.
The cloud is dark grey with bright yellowish-white interstices where
the sun is shining through the thinner parts of the sheet. The

edges of the clouds are fused together; little or no blue sky can be
seen between them.

PLATE 26

CLOUDS: TE7,

PLATE 26.

26a. Strato-cumulus. A high variety of strato-cumulus, almost
as small and fine as alto-cumulus, but this condition existed only at
the edge of an extensive sheet of cloud which was otherwise indubit-
ably strato-cumulus. An edging of smaller cloudlets is a usual
feature in cloud-sheets of this type. The cloudlets in this picture
are thin and flat, almost like small plates or slabs.

268. Strato-cumulus. Part of a long band composed of a series
of very sinuous transverse waves. The dark horizontal band is a
belt of thin stratiform cloud formed just below the strato-cumulus
waves. It is from such films or thin sheets of stratus that occasional
transient showers fall.

118 CLOUDS

PLATE 27.

274. Strato-cumulus. Connected strato-cumulus cloudlets of
somewhat lenticular form exhibiting a wave-form arrangement. It
is worthy of note that the lower band, the one composed of four
cloud-masses, was not in one connected whole when first seen. A
photograph taken a few minutes previously showed the second cloud-
mass from the left to be connected with the now isolated cloud just
below it, but separation from this latter cloud took place simul-
taneously with the connecting-up of the second and third clouds in
the line. A wave-system that had been visible an hour earlier was
thus re-established.

278. Strato-cumulus, Four very straight and well-defined bands
of strato-cumulus of a very heavy type. The bands were moving
towards the observer at right angles to their length, and were
approximately a mile in breadth, the intervening spaces of sky being
also about a mile broad.

FLATE 28

CLOUDS 119

PLATE 28.

28a. Cumulus. Small cumulus clouds at an early stage in their
development. They are photographed with the sun behind the
largest cloud in order to show the characteristic bright edge
surrounding the dark cloud, the appearance which is popularly
known as the “silver lining ”.

288. Cumulus. A very typical cumulus cloud of rather large
dimensions and pronounced shape. The dome-shaped or pyramidal
structure is very evident, the numerous protuberances which indicate
the rising currents may be seen, while the flatness of the dark base,
which marks the level of condensation, is also well shown. Cumulus
clouds of large dimensions, like this one, are those that very
frequently continue growing till they become cumulo-nimbus.

120 CLOUDS

PLATE 20.

29a. Cumulus. A rather unusual arrangement where the
cumulus clouds are grouped into long parallel lines very much like
the parallel waves seen in the higher clouds. ‘The cumulus clouds,
however, were moving in the same direction as that in which the
lines were lying, whereas the waves in the higher clouds usually
travel at right angles or at some other considerable inclination to
their length.

29B. Cumulus (with Alto-cumulus above). A line of connected
cumulus clouds near the coast-line. They appear dark chiefly
because the sun is behind them a little to the right of the picture,
and is also partly obscured by the layer of fused alto-cumulus above.
At Aberdeen it is fairly common for cumulus clouds to form in line
along the coast on winter mornings, when a slight cold westerly
breeze from the land may undercut and elevate the warmer sea air
that may be moving from south or north.

PLATE 2g

PLATE 30

CLOUDS 12I

PLATE 30

30A. Cumulus (and Fracto-cumulus). Small cumulus clouds mixed
with a considerable amount of the broken variety known as fracto-
cumulus. ‘The latter form seems to be accompanied by much turbu-
lence and is found often in squally weather,

308. Cumulus (becoming Strato-cumulus). In this picture
cumulus clouds are shown in the process of flattening out and
assuming the character of strato-cumulus. Air conditions are then
becoming more stable and vertical currents are dying out; the type
is therefore found more frequently towards evening.

122 CLOUDS

PLATE 3r1.

31a. Cumulo-nimbus. A large thunder-cloud with a very flat
base from which rain is falling in the far distance. In the near
front of the cloud-mass the moisture in the rising currents is con-
densing and spreading out into a large “anvil” which is fringed
with false cirrus threads. ar in the rear there may be seen more
false cirrus from other parts of the cloud-mass, which latter seems to
be formed by the union of several cumulo-nimbus clouds.

31B. Cumulo-nimbus. ‘The right front portion of a large thunder-
cloud which shows no “anvil”. The cloud is moving away from
the observer and in its front the white protuberant masses of con-
densation are rising, while in the rear they are falling in rain, and
appear grey in colour. ‘The basal part of the cloud is formed by a
sheet of intensely dark cloud of nimbus type.

PLATE 3r

PLATE 22

CLOUDS 123

PLATE 32.

324. Cumulo-nimbus. An advancing hail-shower cloud that is
practically all “anvil”. Some small growing cumulus may be seen
in front. The finely-formed anvil is topped and surrounded by
false-cirrus filaments. The passage of the cloud overhead was
accompanied by a heavy squall with hail.

32B. Cumulo-nimbus. Another hail-squall cloud. It is moving
from left to right, and showers are falling from it over the sea. As
the cloud moves forward the rising moist air forms in front of it a
line of growing cumulus which lies round the front of the main cloud
like a collar.

124 CLOUDS

PLATE 33.

33a. Cumulo-nimbus. This is an example showing rapidly-
growing vertical columns of cloud that are uniting to form a huge
thunder-cloud, of which this photograph includes less than one-
fourth. A quantity of loose fracto-cumulus is floating round the
main mass.

338. Mammato-cumulus. ‘The appearance often seen under the
base in the rear part of thunder-clouds, and sometimes also seen
widespread on the under surface of a cloud-sheet which resembles
closed strato-cumulus. In the latter case it is quite possible that
the upper surface of the cloud, if it could be seen, would be of the
cumuliform type. Rain is either falling at the time this formation
is visible or it follows soon after. In this picture the cloud substance
is being whirled down and up again by the turbulence of the air-
currents. It is a dangerous cloud for aviators.

CLOUDS 125

PLATE 34.

344. Stratus. ‘This is the really typical form of stratus, a grey
pall of cloud without any detail, and rendered visible only by the
fact that the summit of the distant hill is immersed in the cloud,
which is trailing along its flanks in the ragged masses called fracto-
Stratus.

34B. S/ratus. An evening sky with a sheet of thin lifted fog
passing overhead from seaward in a series of beautiful waves, which,
however, are so thin as to permit the sun’s disc to be seen through
them quite clearly. The cloud greatly resembles strato-cumulus in
its arrangement, but its thinness and the obviously small altitude it
showed proved it to be stratus, and in a very short time the cloud-
layer had become quite uniform normal stratus.

126 CLOUDS

PLATE 35.

35a. Mimbus. Actually a shower falling from the nimbus-base
of a large cumulo-nimbus cloud. It is impossible to photograph the
typical nimbus cloud because the rain, when general, is not visible
to the camera. It is only when an isolated shower shows dark
against the lighter sky beyond that the rain can be shown.

358. Wimbus and a Rainbow. <A screen of rain is falling in the
middle distance, and a primary rainbow, with a faint outer secondary
bow, have become visible. The lighter space within the primary
bow is plainly shown, and near the crest of the arch several super-
numerary bands are seen.

PLATE 36

CLOUDS 127

PLATE 36.

36A. Line-Sguall Cloud. The first stage, showing the beginning
of the formation of the long line of cloud (see p. 63 for description).
This and the three following pictures are from pastel sketches made
by the writer from personal observation.

e

36n. Line-Sguall Cloud. ‘The second stage, where the long line
of cumulus has become complete; the sheet of dark nimbus may be
seen following the squall front. The nimbus sheet was higher than
the line of cumulus, and followed it with a slight interval between
them.

128 CLOUDS

PLATE 37.

37a. Line-Sguall Cloud. The third stage—the line of cumulus
after it had moved out over the sea. The cloud was losing its
cumuliform character and beginning to show some vertical arrange-
ment, while small wisps of condensation were moving up and down
below the cloud front.

«

378. Line-Sguall Cloud. The fourth stage, showing the cloud
front after the formation of the three waterspouts. The cloud was a
whirling turbulent mass with very ragged edges which were drawn
out into thin filaments. Further observation was prevented by the
heavy shower which hid the cloud front from view.

; °

ra ae

a . an

e -

> 7

3 *,,

+ - -
— + 5
“4

. :

L i=

;

:

a

aeerenretina tame.
,
4
:
vt
P
‘

- " ’

ee ae
a A ora Se sate

a

H F
Re Ot i rn 3 ls ip Spe are Gap ess ‘ ie Sen na a renee See. Cee a ST 5 ——s
ee ex. eam a eee LM - - or a 3

‘

Soe 2S at eee

Soke

’

PLATE 38

CLOUDS 129

PLATE 38.

38a. Cumulo-nimbus. A view from above of a massive heap of
cloud which reached from 1,000 to 10,000 feet above the ground,
that is to say, it was nearly two miles in depth. The tops of the
rising air columns are seen remarkably well in this view.

(Aeroplane photograph by Capt. C. K. M. Douglas, R.A.F.)

388. Strato-cumulus. A close view of the upper surface of a
turbulent sheet of strato-cumulus. The top of the sheet was 5,000
feet high, and an inversion of 10° F. occurred above it.

(Aeroplane photograph by Capt. C, K. M. Douglas, R.A.F.)

130 CLOUDS

PLATE 309.

394. Strato-cumulus, ‘The top of a cloud-sheet whose height was
9,700 feet, considerably above the average height of that particular
type of cloud. The cloud billows are very clearly shown, one
system of waves or folds crossing the other at an angle of about
30°. There was an inversion of 3° F. above this layer.

(Aeroplane photograph by Capt. C. K. M. Douglas, R.A.F.)

398. Strato-cumulus. Upper surface of a normal strato-cumulus
cloud-layer, across which there runs a “cliff-front” of cloud. The
height of the surface in the foreground was 3,800 feet, that of the
surface beyond the “cliff-front” was 4,500 feet, there was thus a
sheer rise of 700 feet along the “ cliff” line.

(Aeroplane photograph by Capt. C, K, M. Douglas, R.A.F.)

Gries

eM
"tear

PLATE 40

CLOUDS 131

PLATE 40.

40A. A long narrow strip of cloud rippled into waves which was
formed in a damp layer at 4,000 feet. Below it there are floating a
large number of small cumulus which had formed at 1,000 to 1,500
feet from a sheet of stratus.

(Aeroplane photograph by Capt. C. K. M. Douglas, R.A.F.)

40B. Cumulus. A view taken above a layer of cumulus clouds,
showing three cloud summits which had become detached from
three pillars of condensation that towered upwards from the sheet
below. The intervening parts of the cloud-pillars had evaporated
again.

(Aeroplane photograph by Capt. C. K. M. Douglas, R.A.F.)

INDEX.

A

Abercromby, Hon. Ralph, 5.

Adiabatic lapse-rate, 57, 58, 59, 60
pie

Aerial navigation and cloud, 81.

Aeroplane observation of clouds, 6,
26, 35, 70.

Alto-cumulus, 8, 10, 25-26, 27, 77,

86, 87.

— castellatus, 28.

— — sporadic occurrence of, 29.

— lenticular, 29-33.

Alto-stratus, 8, 10, 27, 33-34, 77, 87,
89, 90.

Angot, A., 75, 83.

Anticyclone, 54, 72, 84- -90.

“Anvil” of cumulo-nimbus, 39, 41,
43, 66, 86. |

Arcs of contact to halos, 21. .

Ascent of air; 38, 53, 62, 66, 71, 72)
(see also ‘“‘ Convection”).

Association of cloud and weather, 33,
69, 83-90. :

Atlas of clouds, International, 5.

Atmospheric equilibrium, 62, 66.

B

Balloon ascents, 46, 58, 59, 65, 70,
71, 76.

Belt of maximum cloudiness, 75.

Bigelow, F. H., 76.

Bjerknes, V., 53 (footnote).

Cc

Causes of clouds, 51-74.

Cave, Capt. C. J. P., 18, 26, 28, 32,

38, 77.

Cirro-cumulus, 8, 9, 18

48, 69, 77, 83, 87.
— lenticular, 29-33.

23-25, 34,

, , Cirro-macula, 23.

Cirro-nebula, 22.

Cirro-stratus, 8, 9, 17, 19-23, 34, 77;
87, 88, 89, go.

Cirrus, 5) 8, 9, 15-19, 69, 79; 81, 83,
87, 89.

Cirrus-haze, 22.

Classifications, cloud, 5, 6.

Clayden, A. W., 6, 18, 24, 28, 37.

Clearing squall, go.

“ Cliff-like ” structure
cumulus, 35.

Cloud Atlas, International, 5.

composition of, 18, 25, 26, 66, 72.

directions, 76.

distribution, 75.

frequencies, 80.

heights, 8, 9, 14, 36, 46, 76, 79,
80.

layers, VOmios

— motions, 76.

Cloudiness, 75.

Clouds and aerial navigation, 81.

— — weather, 33, 69, 83-90.

— formation of, 51-74.

Colours in rainbows, 44.

— of the clouds, 33, 37, 47- a

in strato-

Composition of clouds, 18, 25, 26,
66, 72.
| Condensation, Aan 2. Ol Ob sa71Os

89.
Conduction, 53, 58.
Convection, 38, 56, 58, 60, 62, 66.
Cooling of the air, modes of, 52, 68.
Corless, R., 63.
Corone, 25.
Counter-sun, 21.

(133)

134

Cumulo-nimbus, 9, 11, 37-45, 56, 62,
66, 78, 79, 80, 82, 86, go.

Cumulus, 5,9, II, 35, 37-455 56, 62,
65, 66, 78, 79, 80, 82, 84, 86, 87,
89, 90.

Cyclone, 54, 73; 83, 84-90.

D

Daily weather report, 81.
Dappled sky, 26.
Depression, 53, 54, 73, 77, 83, 84-

go.
Depth of clouds, 14, 37, 72.
Descent of air, 38, 65, 71.
Diffraction, 25, 44, 50.
Dines, W. H., 58, 72.
Directions of clouds, 76.
Distribution of cloudiness, 75.
— — pressure, 53, 84-90.
Diurnal range of cloud, 42.
Douglas, Capt. C. K. M., 26, 35, 70,
71, 72.
Dynamical cooling of air, 53.

E

Eddy-conductivity, eddy-motion, 67,
69, 71, 84.

Edges of cloud-layers, sharp, 72, 9o.

Effect of perspective, 15, 37.

Elevation of air, 53, 65, 71, 72 (see
also ‘* Convection ”’).

Equilibrium, atmospheric, 62.

Evaporation, 51, 52.

Expansion of air, 53, 58, 60, 69.

)e:

Fall cloud, 33.

— of pressure, 53, 69, 72.

— — temperature with height, 56.
False-cirrus, 11, 12, 26-28, 34, 39.
Fog, 5, 45, 68.

Forced rising of air, 53.
Formation of clouds, 51-74.
Fracto-cumulus, 11, go.
Fracto-nimbus, 11, go.
Fracto-stratus, 12, 86.
Frequencies of clouds, 80.

CLOUDS

G

Geddes, Capt. A. E. M., 55, 65.
“« Geophysical Journal,” 77.
Gold, Lt.-Col. E., 58.

Gradient, barometric, 54.

— vertical temperature, 56, 57.

H

/
Hail, 38, 39, 42.
Halos, 19, 22.
Height of stratosphere, 59.
Heights of clouds, 8, 9, 14, 36, 46,
76, 79, 80.
High-pressure areas, 54, 72, 84-90.
Hildebrandsson, H., 5, 79.
Howard, Luke, 4, 5.
Humidity, 61, 70, 71.

I
Ice-crystal clouds, 18, 19, 22, 26, 27
66.

International classification of clouds,
5, 7, 8.

— Cloud Atlas, 5.

— cloud observations, 76, 79.

Inversions, 58, 62, 67, 68, 71.

Ionised air and condensation, 51.

Iridescent clouds, 50.

Irisation, 50.

Isobars, 53, 54, 84.

Isothermal layer, 59.

K
Kite ascents, 71.
|

Lapse-rates, 57, 58, 59, 60, 61, 71.

Lempfert, R. G. K., 55, 63.

Lenticular clouds, 12, 29-33, 35, 47,

48, 50, 74, 87.

— openings in cloud-sheet, 30.

Ley, W. Clement, 5, 23, 28.

Line-squalls, 42, 56, 62-66.

Local influences on clouds, 32, 65,
66, 74, 84.

Low-pressure area, 53, 56, 84-90
(see also “ Cyclone ”’),

INDEX

M

Mackerel sky, 26.

Mammato-cumulus, 12, 39, 90.

Mamumillation, 34, go.

Maximum of cloud, 76, 80.

Meteorological Office, 28, 81.

Minimum of cloud, 76, 80.

Mixing of air, 53, 70.

Mock sun ring, 21.

Moon pillar, 22.

Mountain stations, cloud at, 76.

Mountains, influence of, on cloud,
12, 32, 74, 84.

.N
Nephoscopic observations of cloud,

77:
Newfoundland Banks, fog on, 68.
Nimbus, 8, 11, 35-37, 53, 56, 69, 79,
84, 87, 89, 90.
North Sea fog, 68.

0)

Observations of cloud at Aberdeen,
19, 24, 27, 29, 33,
39) 41, 49, 93, 73,
77, 79, 80, 84.

— Epsom, 8o.

— meteorological
tions, 81.

in Devon, 24.

— early times, 3.

— Hampshire, 32.

Optical phenomena, 19-22, 25, 43,

50.
Ovoid-cloud form, 12, 30.

sta-

P

Paraselene, 21.

Parhelia, 21.

Paths of air, 55.

Periodicity of cloud, 80.

Persistence of cloud, 46.

Perspective, effect of, 15, 37.

Phenomena, optical, 19-22, 25, 43,
50.

Pressure, fall of, 53, 69, 72.

— distribution of, 53, 84-90.

135
R

Radiant-point, 17.

Radiation, 53, 58, 70, 72.

Rainbows, 43.

Rain-cloud (see ‘“‘ Nimbus ”).

Rainfall and cloud, 53, 56, 65 (see
also ‘‘ Nimbus ny?

Bevis of cloud-formation, 23,

9.
Rate of fall of temperature with

24,

height, 57 (see also ‘ Lapse-
rates ”).

Refraction, 19, 44.

Reticulated structure, 22, 34, 77.

Riggenbach, A., 5.

Ripples in clouds, 73.

Rising of air (see ‘Ascent of Air”
_ and “ Convection ”).

Ruskin, J., 4, 15, 17, 26, 49.
Russell, S. C., 80.

Ss

Saturation, 51.

teSeud Zecloudy 1114930, 075) OO}

Secondary depression, 9o.

Sequence of cloud in depression, 83,
89.

Sharp edges to cloud-layers, 72, 90

Shaw, olre Napier, 275 31,55) 58)
63, 69, 84, OI.

Shearing plane between air-currents,
73-

Shimmering, atmospheric, 68.

Shower-cloud (see “ Cumulo-nim-
bus”).

Simpson, G. C., 50, 66.

Size of clouds, 14, 37.

Speckle cloud, 23.

Squalls, 42, 56, 62, 66, 67.

Stationary nature of lenticular clouds,

3D.

— waves in atmosphere, 74.

Straight edges to cloud-sheets, 72,90.

Strato-cumulus, 8, 10, 34-35, 71, 77,
80, 84, 86, 87, 89, 90.

Stratosphere, 59, 69, 79.

Stratus, 5, 9, 12, 45-47, 69, 79, 84,
86, 87.

Sun-pillar, 22.

136

Sunset colours, 47-48.
Supercooled water-drops, 19, 66.
Supernumerary rainbows, 44.
Super-saturation, 51.

ih

Taylor, Major G. I., 68.

Teisserenc de Bort, L., 5, 79.

Telescopic observations of clouds,
73-

Temperature gradient, vertical, 56,

57+
Thickness of clouds, 14, 36, 37, 52,
Tee
Thunder-cloud (see ‘“Cumulo-nim-
bus”).

Poe and _ associated
_ clouds, 29, 39, 41, 47, 63, 90.
Trajectories of air, 55.

Troposphere, 59.

Trough line of depression, 73, 89.
Turbulence of flowing air, 53, 67.
— in thunder-clouds, 38, 66.
Turner, J. M. W., 4. —
Turret-cloud, 28.

— sporadic occurrence of, 29.

U

Undercutting of warm air by cold,
62, 65, 66, 72.

CLOUDS

“Upper Air Supplement to Daily
Weather Report,” 81.

Vv

V-shaped depression, go.
Vanishing-point, 17.

Variation in cloud amounts, 75.

— — — heights, 76.

Vector diagrams of cloud-motion,

Velocities of clouds, 79, 80.
Vertical temperature gradient, 56,

57
Visibility, 45, 86.

Ww

Waterspouts, 65.
Water-vapour, 51, 52.

Watery sky, 34.

Waved clouds, 73, 74.
Weather and clouds, 33, 69,

go.
Weather-type, 83-90.
Whipple, F. J. W., 44.
Whirlwinds in line-squalls, 65.
Winds, weather, and cloud, 54, 63,
76, 84-90.

ABERDEEN; THE UNIVERSITY PRESS

ee
*
ns

ee Be ey ei ee

>

SSS e5
es SePreSeRE ESSA SEES SERS
ifs Bt artereerireretar iy

isis

i ; He

+

ig
53955599 3993